Lawyers, Interrogations and the Historic Framework of ...€¦ · 2010] LAWYERS, INTERROGATIONS AND THE HISTORIC FRAMEWORK 103 personalities, or they are sources of information about

Lawyers, Interrogations and the Historic Framework of Debate About Torture

SHELDON GELMAN †

I. INTRODUCTION

Recent controversy about torture and interrogation has centered on formal opinions by the Office of Legal Counsel (hereinafter “OLC”) during the Bush administration.1 Construing the anti-torture statute narrowly, these opinions upheld simulated drownings, painful stress positions, extended isolation, and slamming detainees into specially constructed walls, among other methods.2 Conceded by OLC to be harsh, many believe these methods constitute illegal torture.

† Professor of Law, Cleveland-Marshall College of Law. Research support was provided by the Cleveland-Marshall Fund. I am also grateful to James Wilson and Brian Ray for their comments and advice.

1 See, e.g. Memorandum from Jay S. Bybee, Assistant Att’y Gen., Office of Legal Counsel, to Alberto R. Gonzales, Counsel to the President, on Standards of Conduct for Interrogation Under 18 U.S.C. §§2340–2340A (Aug. 1, 2002) [hereinafter Torture Memo], in THE TORTURE PAPERS: THE ROAD TO ABU GHRAIB 172 (Karen J. Greenberg & Joshua L. Dratel eds., 2005) [hereinafter THE TORTURE PAPERS].

2 18 U.S.C. §§ 2340–2340A (2006), which provides: § 2340. Definitions As used in this chapter- (1) “torture” means an act committed by a person acting under the color of

law specifically intended to inflict severe physical or mental pain or suffering...; (2) “severe mental pain or suffering” means the prolonged mental harm

caused by or resulting from- (A) the intentional infliction or threatened infliction of severe

physical pain or suffering; (B) the administration or application, or threatened

administration or application, of mind-altering substances or other procedures calculated to disrupt profoundly the senses or the personality;

(C) the threat of imminent death; or (D) the threat that another person will imminently be subjected

to death, severe physical pain or suffering, or the administration or application of mind-altering substances or other procedures calculated to disrupt profoundly the senses or the personality . . . .

See also Torture Memo, supra note 1; Memorandum from Jay S. Bybee, Assistant Att’y Gen., Office of Legal Counsel, to John Rizzo, Acting Gen. Counsel of the Cent. Intelligence Agency, on Interrogation of al Qaeda Operative (Aug. 1, 2002) [hereinafter Qaeda Operative Memo], in THE SECRET TORTURE MEMOS: BUSH ADMINISTRATION MEMOS ON TORTURE AS RELEASED BY THE UNITED STATES DEPARTMENT OF JUSTICE APRIL 16, 2009, (2009) [hereinafter THE SECRET TORTURE

102 CONNECTICUT PUBLIC INTEREST LAW JOURNAL [Vol. 10:1

A. The Historic Framework of Controversy

The debate has unfolded within a centuries-old framework, dating back to authors such as Beccaria and Voltaire.3 Following the historic pattern, arguments against torture posit abstract entities—personal and national identity, humanity, civilization, the rule of law—with complex internal structures and inherent value. Critics of torture say that these entities are disregarded, and then destroyed, by torture. Because both torturers and victims can possess these qualities, both are injured. Torture assaults victims, but it also destroys the torturers’ own humanity. Torture produces knowledge, the critics say, but it is primarily knowledge about the character of the torturers.

The other side of the debate employs a different ontology. These arguments focus on material things and physical survival or destruction, not on abstractions. Neither the victims nor the torturers have a complex internal life (or if they do, it is not mentioned). The focus is instead on singular events, which become subjects of calculation and measurement—if this bomb goes off, many people will die.

On one view, a nation that employs torture threatens its own identity.

On the other view, the nation is protecting itself against destruction of its people and property. Torture victims are either persons with distinctive

MEMOS]; Memorandum from John C. Yoo, Deputy Assistant Att’y Gen., Office of Legal Counsel, to William J. Haynes II, Gen. Counsel of the Dep’t of Def., on Military Interrogation of Alien Unlawful Combatants Held Outside the United States (Mar. 14, 2003) [hereinafter Military Interrogation Memo], available at www.justice.gov/olc/docs/memo-combatantsoutsideunitedstates.pdf; Memorandum from Daniel Levin, Acting Assistant Att’y Gen., Office of Legal Counsel, to the Deputy Att’y Gen. on Legal Standards Applicable Under 18 U.S.C. §§2340–2340A (Dec. 30, 2004) [hereinafter Legal Standards Memo], available at http://www.justice.gov/olc/18usc23402340a2.htm; Memorandum from Steven G. Bradbury, Principal Deputy Assistant Att’y Gen., Office of Legal Counsel, to John A. Rizzo, Senior Deputy Gen. Counsel, Cent. Intelligence Agency, on Application of 18 §§2340–2340A to Certain Techniques That May Be Used in the Interrogation of a High Value al Qaeda Detainee (May 10, 2005) [hereinafter Techniques Memo], in THE SECRET TORTURE MEMOS, supra at 29; Memorandum from Steven G. Bradbury, Principal Deputy Assistant Att’y Gen., Office of Legal Counsel, to John A. Rizzo, Senior Deputy Gen. Counsel, Cent. Intelligence Agency, on Application of 18 §§2340–2340A to the Combined Use of Certain Techniques in the Interrogation of High Value al Qaeda Detainees (May 10, 2005) [hereinafter Combined Techniques Memo], in THE SECRET TORTURE MEMOS, supra at 77; Memorandum from Steven G. Bradbury, Principal Deputy Assistant Att’y Gen., Office of Legal Counsel, to John A. Rizzo, Senior Deputy Gen. Counsel, Cent. Intelligence Agency, on Application of United States Obligations Under Article 16 of the Convention Against Torture to Certain Techniques That May Be Used in the Interrogation of High Value al Qaeda Detainees (May 30, 2005) [hereinafter Article 16 Memo], in THE SECRET TORTURE MEMOS, supra at 99.

3 See Christopher Kutz, Torture, Necessity and Existential Politics, 95 CAL. L. REV. 235, 239 (2007) (describing the “eradication of the moral and legal basis for torture” as “one of the defining features of post-Enlightenment liberal politics” that traces back to “early polemics of Voltaire and Beccaria”); ALFRED W. MCCOY, A QUESTION OF TORTURE: CIA INTERROGATION, FROM THE COLD WAR TO THE WAR ON TERROR 14 (2006) (noting that over “the past two centuries . . . repudiation [of torture] has been synonymous with the humanist ideals of the Enlightenment and democracy.”).

2010] LAWYERS, INTERROGATIONS AND THE HISTORIC FRAMEWORK 103

personalities, or they are sources of information about a particular thing, such as the location of a bomb.

This apparently simple argumentative dichotomy has surprising power. Author after author conforms to its strictures. Thus, the current controversy concerns more than the effectiveness of interrogations and the amount of harm they entail. It is also—and perhaps primarily—about the kind of entities and the types of damage that deserve our attention.

This pattern of argument is not compelled by logic. Arguments for torture (or harsh interrogation) could cite abstract entities. Indeed, an important argument for torture once did, positing that torture benefitted the soul.4 More recently, abstract ideas about race and ethnicity have been cited to justify the exclusion of persons from the national or human community, making their suffering under torture or other oppressions seem morally acceptable and perhaps even desirable. The fact that such arguments now seem barbaric does not make them any less abstract.5

B. Making Moral Judgments

OLC’s opinions on interrogation have been criticized ethically and morally, not just legally. However, OLC attorneys were not responsible for the torture statute. If the statute permitted immoral torture, the responsibility lies with Congress. Without penetrating the “mask of the law,” then, moral and ethical judgments about the opinions seem unwarranted.

Once public, the interrogation opinions produced a firestorm of criticism.6 Newspaper editorialists and commentators often presumed that OLC’s interpretations were too morally abhorrent to be correct.7 Many

4 See, e.g., Andrew Sullivan, The Abolition of Torture: Saving the United States from a

totalitarian future, THE NEW REPUBLIC (Dec. 19, 2005) available at http://www.tnr.com/article/ politics/the-abolition-torture-0 (discussing the use of torture during Europe’s religious wars of the sixteenth and seventeenth century as a means of “[saving] the victim’s soul”). Such torturers posited that the only way to reach the depths necessary for religious conversion “was to deploy physical terror in the hopes of completely destroying the heretic’s autonomy. . . . [Torturers] would, in other words, destroy a human being’s soul in order to save it.” Id.

5 It should be emphasized that arguments of this kind were never made by OLC. 6 See David Margolis, Associate Deputy Att’y Gen., Memorandum of Decision Regarding the

Objections to the Findings of Professional Misconduct in the Office of Professional Responsibility’s Report of Investigation into the Office of Legal Counsel’s Memoranda Concerning Issues Relating to the Central Intelligence Agency’s Use of “Enhanced Interrogation Techniques” of Suspected Terrorists 3 (Jan. 5, 2010) [hereinafter Margolis Opinion”], available at http://judiciary.house.gov/hearings/ pdf/DAGMargolisMemo100105.pdf (recounting newspaper leaks of the Torture Memo).

7 See, e.g., Andrew Rosenthal, Legal Breach: The Government’s Attorneys and Abu Gharib, N.Y. TIMES, Dec. 30, 2004, at A22 (arguing that attorneys in the Bush administration employed “Strangelovian logic” in order to uphold “methods that clearly violated the Geneva Conventions and anti-torture statutes”); Robert Scheer, Tout Torture, Get Promoted?, L.A. TIMES, June 15, 2004,


legal authorities found OLC’s work grossly substandard, marred by serious technical errors.8

Sharing the moral and political outrage, some legal critics also recognized the existence of legal ambiguity, vagueness, and indeterminacy in the anti-torture statute, which employed broad terms such as “severe,” “prolonged” and “suffering.”9 These critics found that the flaws in OLC’s analyses reflected deeper problems in legal philosophy and approach.10

A preliminary July 2009 OPR Report found analytical shortcomings in the opinions serious enough to violate a lawyer’s ethical “duty to exercise independent legal judgment and render thorough, objective, and candid legal advice.”11 This preliminary finding was rejected by a higher level reviewer in the Justice Department, however. The reviewer, David Margolis, agreed that the opinions “contained some significant flaws.”12 However, he also determined that some analysis that OPR had faulted was arguably correct, and that almost all the errors were more debatable than OPR believed.13 Margolis even implied that some of OPR’s own analyses suggested an effort to reach a predetermined legal conclusion—the very available at http://articles.latimes.com/2004/jun/15/opinion/oe-scheer15 (arguing that Jay Bybee’s “standard” for defining torture “should bring comfort to the totalitarian governments that find the brutal treatment of prisoners a handy tool in . . . fighting wars”); Editorial, Yes, It Was Torture, and Illegal, N.Y. TIMES, Jan. 4, 2010, at A16 [hereinafter Torture Editorial] (“Bush administration officials came up with all kinds of ridiculously offensive rationalizations for torturing prisoners.”).

8See, e.g., Confirmation Hearing on the Nomination of Alberto R. Gonzales to be Attorney General of the United States Before the S. Comm. on the Judiciary, 109th Cong. 158 (2005) (statement of Harold Hongju Koh, then Dean, Yale Law School) (describing the Torture Memo as “perhaps the most clearly erroneous legal opinion I have ever read”); Adam Liptak, Legal Scholars Criticize Memos on Torture, N.Y. TIMES, June 25, 2004, at A14 (quoting Cass Sunstein, a law professor at the University of Chicago, who referred to the interrogation opinions as “very low level, . . . very weak, embarrassingly weak, just short of reckless”). See also W. Bradley Wendel, Legal Ethics and the Separation of Law and Morals, 91 CORNELL L. REV. 67, 68 & n.2 (2005) (observing that “[t]he overwhelming response by experts in criminal, international, constitutional, and military law was that the legal analysis in the government memos was so faulty that the lawyers’ advice was incompetent”); Kutz, supra note 3, at 241–48 (detailing errors in OLC opinions); DEP’T OF JUSTICE, OFFICE OF PROF’L RESPONSIBILITY, REPORT: INVESTIGATION INTO THE OFFICE OF LEGAL COUNSEL’S MEMORANDA CONCERNING ISSUES RELATING TO THE CENT. INTELLIGENCE AGENCY’S USE OF “ENHANCED INTERROGATION TECHNIQUES ON SUSPECTED TERRORISTS” 2–4, 8 (July 29, 2009), [hereinafter OPR Final Report] available at http://judiciary.house.gov/hearings/pdf/OPRFinalReport090729.pdf (reviewing the reaction to the Torture Memo from “members of the legal community”).

9 See 18 U.S.C. §§ 2340-2340A (2006). 10 Jeremy Waldron, Torture and Positive Law: Jurisprudence for the White House, 105 COLUM.

L. REV. 1681, 1691 (2005) (suggesting that the basic mistake in legal analyses was applying standard “strategies of [legal] interpretation” when the torture issue required something else entirely); Kutz, supra note 3, at 244, 275 (noting that “the Bybee memo looks as though it stakes out a defensible legal position” on the necessity defense but that the memo’s argument fails for philosophical reasons).

11 OPR Final Report, supra note 8, at 11. 12 Margolis Opinion, supra note 6 at 67. 13 See generally id. at 28–63 (reviewing alleged shortcomings in the opinions).


failing OPR identified in the interrogation opinions themselves.14 He concluded that while the interrogation opinions failed to meet “the Department’s high expectation of its OLC attorneys,” they fell short of violating rules of professional conduct.15 According to Margolis, interpretations of the torture statute seemed so debatable that it was difficult to be blatantly mistaken.16

However, some legal critics already had identified a flaw beyond any simple misreading of statutory or constitutional language. These critics saw a fundamentally misguided interpretive approach at OLC, one that ignored important principles.17 Slighting such principles, OLC attorneys selected the wrong statutory meanings from the array of dictionary alternatives, with disastrous consequences. While different legal critics invoked different principles, they all perceived OLC’s interpretative error as arising from a mistake about fundamentals. And this mistake, which went well beyond any misreading of the torture statute, was said to warrant moral condemnation.

There remained a problem, however. It was one thing for OLC to commit glaring legal errors; however, because highly qualified lawyers had produced the opinions, such errors were unlikely to arise from purely technical causes.18 Experienced professionals do not repeatedly commit blatant errors without some purpose. Thus, the best explanations for OLC’s errors seemed to involve political pressure from the administration and the attorneys’ personal ambitions and desire to please the White House. Such pressures prompted lawyers to say what the administration wanted and to sanction torture, lapses that clearly deserved ethical and moral censure.

Thus, lay critics stripped away the mask of law using blatant legal error as the tool. But the more susceptible that statutory language is to different readings, the harder it becomes, as we have seen, to find a

14 Id. at 8 (quoting John Yoo’s argument that an earlier draft of the OPR Final Report “goes to

great lengths to criticize what it asserts was ends-driven legal reasoning” in the Torture Memo when “OPR has itself engaged in exactly this alleged sin”). See generally id. at 7–26 (discussing changes in OPR’s reasoning during successive report drafts).

15 Id. at 68. 16 Margolis Opinion, supra note 6, at 28–35. 17 E.g., Wendel, supra note 8, at 82 (attributing technical errors in the opinions to “OLC lawyers’

[general] mode of analysis, which is to rely on formalistic and narrow constructions of legal rules, divorced from their context and other sources of meaning”); Kutz, supra note 3, at 238 (arguing that the opinions “betray a failure to grasp . . . [the] distinction[]” between kinds of rights and a failure to understand the nature of “core human rights protections”).

18 See, e.g., Wendel, supra note 8, at 70 ( “It is difficult to credit the explanation that the authors . . . [of the opinions] were incompetent, since they worked for agencies—such as the OLC—which traditionally employ some of the very best legal talent in the country.”).


particular interpretation blatantly wrong. That, however, undercuts the argument that OLC lawyers acted unethically or immorally, as Margolis’ overturning of OPR’s ethical conclusions demonstrated.19 Predictably, OPR lawyers—and, in its later opinions, OLC itself—emphasize the vagueness and indeterminacy of language in the anti-torture statute.20

In place of technical but ephemeral errors, legal writers have cited OLC’s traducing deeper interpretative principles, for example, the principle that lawyers “must treat legal norms as legitimate reasons for action in their practical deliberation”21 or that a prohibition of torture “operates . . . as an archetype . . . —as a rule which has significance not just in and of itself.”22 No matter how elegantly or carefully those principles are presented, however, they afford a more questionable basis for ethical or moral criticisms of OLC than blatant legal errors would. Indeed, each of these principles is itself debatable.

Here, the historic framework is instructive. Each proposed interpretive principle invokes an abstract idea of law. As demonstrated below, however, OLC’s interpretations were emphatically anti-abstraction—a circumstance that aligns its interpretations with classic pro-torture arguments. It identifies OLC’s supposedly neutral legal approach with a recognizable historical position.

The framework approach avoids the problematic assumptions made by lay critics. It neither ignores legal indeterminacy nor overlooks the distinction between law and morals. In addition, the fact that OLC lawyers may have sincerely held their legal views23—even apparently mistaken views—proves problematic for conventional ethical and moral criticisms, but not for analyses under the framework. Sincerity undercuts the

19 An extreme degree of legal indeterminacy would make it easier, not harder, to draw moral inferences. If statutory language imposed no restraint on the possible interpretations, law would mean whatever the interpreter wanted it to mean: no legal interpretation would be wrong, and every interpretation seemingly would reflect a moral judgment. Since OLC’s critics identify legal errors in its work, they clearly do not subscribe to this extreme version of legal indeterminacy.

20See JOHN YOO, WAR BY OTHER MEANS: AN INSIDER’S ACCOUNT OF THE WAR ON TERROR 171 (2006) (noting that the statute “used words rare in the federal code” and that “it had never been interpreted by a federal court”); JACK GOLDSMITH, THE TERROR PRESIDENCY: LAW AND JUDGMENT INSIDE THE BUSH ADMINISTRATION 143 (2007) [hereinafter THE TERROR PRESIDENCY] (noting that the meaning of provisions of the anti-torture statute are “not obvious”); Techniques Memo, supra note 2, at 73 (noting that “reasonable persons may disagree” about the issues and that interpreting the anti-torture law “has been made more difficult by the imprecision of the statute and the relative absence of judicial guidance”).

21 Wendel, supra note 8, at 72. 22 Jeremy Waldron, supra note 10, at 1687 (2005) (emphasis in original). 23 Margolis Memo, supra note 4, at 67 (“Yoo’s loyalty to his own ideology and convictions

clouded his view of his obligations to his client and led him to author opinions that reflected his own extreme, albeit sincerely held, views.”). See also GOLDSMITH, supra note 20, at 97–98 (discussing Yoo’s personal beliefs regarding executive power and how they factored into his OLC memoranda).


inference that a lawyer yielded unprofessionally to pressure or shaped his or her legal views to suit a superior. Using the framework as an analytical device, however, sincerity presents no problem; whether views are held sincerely or not, the framework can establish their moral genealogy.

C. The OLC in the Bush Administration

OLC’s history in the Bush administration falls into three periods. In 2002, Jay Bybee headed OLC, and John Yoo served on the agency’s staff. On August 1, 2002, OLC issued two interrogation opinions: the “Torture Memo,” a general analysis of the anti-torture statute and related law, and the “Qaeda Operative Memo,” which upheld waterboarding, extended stress positions, walling, sensory deprivation, and other methods in the case of a particular detainee.24

Drafted by Yoo and signed by Bybee,25 the Torture Memo construed the anti-torture statute to require a level of pain commensurate with organ failure or death before an interrogation technique could constitute torture. It also concluded that Congress was powerless to limit the President’s power to command “battlefield” interrogations; that the statutory requirement of “specific intent” to engage in torture was exceedingly hard to satisfy; and that defendants prosecuted for violations of the anti-torture statute had available broad defenses of “necessity” and “[national] self-defense.”26 When it leaked to the press, the Torture Memo became the flashpoint for criticism of Bush administration policies.

The second period began in the Fall of 2003, when Bybee received a Court of Appeals appointment. Yoo was passed over for promotion, and Jack Goldsmith became OLC director.27 A former member of the University of Chicago law faculty, Goldsmith had specialized in national security law and international relations.28 Soon after being appointed OLC director, he clashed with representatives of the White House and Vice President Cheney’s office; among the matters in dispute was the Torture Memo, which Goldsmith considered flawed.29 After a stormy nine month

24 Torture Memo, supra note 2; Qaeda Operative Memo, supra note 2. 25 OPR Final Report, supra note 8, at 251 (describing Yoo as “directly responsible for the

contents”); id. at 255 (describing Bybee’s role as signator of the opinions). 26 Torture Memo, supra note 1, at 174, 207. 27 OPR Final Report, supra note 8 at 110 n.83. Jeffrey Toobin, Hiding Jay Bybee, THE NEW

YORKER, (Apr. 16, 2009), http://www.newyorker.com/online/blogs/newsdesk/2009/04/jeffrey-toobin-hiding-bybee.html.

28 GOLDSMITH, supra note 20, at 20–21. Goldsmith’s first position in the Bush Administration was in Pentagon where he served as “Special Counsel” to James Haynes. Id.

29 Id. at 41, 78–79, 151.


tenure, Goldsmith formally withdrew the Memo and resigned.30 However, Goldsmith did not produce any formal written opinions on torture during his tenure,31 and he never concluded that the interrogation techniques sanctioned by Bybee and Yoo were illegal.32 Three years later, after joining the Harvard Law School faculty,33 Goldsmith published a memoir of his time at OLC entitled The Terror Presidency: Law and Judgment Inside the Bush Administration.34

The third period extends from Goldsmith’s 2004 departure from OLC through the end of the Bush Administration. OLC issued formal opinions during this period to replace those produced earlier. Written more narrowly—avoiding the issues of inherent presidential power and of possible criminal defenses; eschewing the “pain equivalent to organ failure” definition of torture; and somewhat moderating the understanding of “specific intent”—these opinions nonetheless upheld the same interrogation methods as Bybee and Yoo’s memos had done. This chapter in OLC’s history ended with the change of administrations, when President Obama renounced use of the enhanced interrogation methods.35

D. Jack Goldsmith

The Terror Presidency is a lucid and engaging account. Goldsmith’s book effectively strips away the mask of law from OLC’s thinking. What it reveals, however, conforms to the classic pattern of pro-torture arguments.

Many of Goldsmith’s views in The Terror Presidency seem to be representative of the OLC generally. Goldsmith himself says that he belongs in the OLC mainstream.36 Some apparently striking arguments that he makes in the book—that OLC’s legal conclusions should be influenced by the President’s “agenda,”37 for example, or that OLC

30 Id. at 160–61. 31 Id. at 158, 162. 32 Id. at 153. Goldsmith did, however, raise questions about whether field agents were going

beyond the OLC limits and he recommended suspension of waterboarding pending further review. OPR Final Report, supra note 8, at 115.

33 GOLDSMITH, supra note 20, at 161. 34 Id. 35 Exec. Order No. 13,491, § (3)(b), 74 Fed. Reg. 4893–94 (Jan. 27, 2009). See also William

Ranney Levi, Note, Interrogation’s Law, 118 YALE L. J. 1436, 1454–55 (2009) (discussing President Obama’s action).

36 GOLDSMITH, supra note 20, at 34–35, 38 (discussing the views of past OLC directors, from both political parties).

37 See discussion infra Part IV.B.2.


resembles an autonomous court for the executive branch38—also appear in a 2004 statement of “Principles to Guide the Office of Legal Counsel” signed by nineteen former OLC lawyers in both Republican and Democratic administrations.39 Regarding interrogation in particular, Goldsmith observes that when he withdrew the Torture Memo, almost every senior administration lawyer agreed with him about its flaws.40 Moreover, Goldsmith specifically endorses the later OLC opinion that replaced the Torture Memo;41 indeed, he supervised early work on that document.42 It is true that Goldsmith and Yoo parted company over interrogation and other matters, but, as the OPR reports make clear, Goldsmith, not Yoo, represents the historic OLC norm.

My assessments of Goldsmith and his book are at odds with some widely held opinions. In 2006, for example, a national news magazine described Goldsmith as the leader of a “palace revolt” who “demand[ed] . . . the White House stop using . . . farfetched rationales for riding rough-shod over the law and the Constitution” and who “fought to bring government spying and interrogations methods within the law.”43 These efforts “did not always succeed,” according to the report, but they “went a long way toward vindicating the principle of a nation of laws and not men.”44 This account, and many like it, effectively assigns Goldsmith to the anti-torture side of the historic debate. Yet Goldsmith’s actual positions conform far more closely to the pattern of pro-torture arguments. Far from seeing brutal interrogations as a threat to the rule of law, Goldsmith considered law and lawyers a threat to Bush administration

38 See discussion infra Part IV.B.3. 39 Memorandum from Walter E. Dellinger et. al., Principles to Guide the Office of Legal Counsel

(December 21, 2004), reprinted in OPR Final Report, supra note 8, at attachment F. Among these guiding principles was number four: “OLC’s legal analyses, and its processes for reaching legal determinations, should not simply mirror those of the federal courts, but also should reflect the institutional traditions and competencies of the executive branch as well as the views of the President who currently holds office.” Id. at 3 (emphasis in original).

40 GOLDSMITH, supra note 20, at 161. 41 Id. at 164–65 (internal citations omitted) (noting that “no approved interrogation technique”

was disapproved by the Legal Standards Memo and that the errors of the Torture Memo were therefore “completely unnecessary to the tasks at hand”).

42 OPR Final Report, supra note 8, at 123 n.93. 43 Daniel Klaidman et al., Palace Revolt, NEWSWEEK, Feb. 6, 2006, at *1, available at

http://www.newsweek.com/2006/02/05/palace-revolt.html. 44Id. A similar account appears in JANE MAYER, THE DARK SIDE: THE INSIDE STORY OF HOW

THE WAR ON TERROR TURNED INTO A WAR ON AMERICAN IDEALS 281–94, 334–35 (2008). But cf. Scott Shane and David Johnston, Lawyers Agreed on Legality of Brutal Tactic, N.Y. TIMES, June 7, 2009, at A1 (“None of the Justice lawyers who reviewed the interrogation question [including Goldsmith] argued that the methods were clearly illegal.”).


policies in the “War on Terror.”45 In fact, Goldsmith did not withdraw the interrogation opinions because

of any doubts about the legality of interrogation methods that OLC had specifically approved. While Goldsmith did believe that the Torture Memo included legal errors, these struck him as of secondary importance; what he principally objected to was Yoo’s embrace of broad legal principles and assertions.46

Thus, Goldsmith singles out for praise Yoo’s Qaeda Operative Memo of August 2, 2002, which upheld all the administration’s enhanced interrogation methods—including waterboarding—in the case of a particular detainee, Abu Zubaydah. It was “deeply strange,” Goldsmith told OPR investigators, that the Torture Memo and the Qaeda Operative Memo bore the same date because:

One . . . [the Qaeda Operative Memo] is hyper narrow

and cautious and splitting hairs and not going one millimeter more than you needed to answer the question. And the other . . . [the Torture Memo] is the opposite. It wasn’t addressing particular problems. It was extremely broad. It went into all sorts of issues that weren’t directly implicated . . . .47

The Terror Presidency itself pays little attention to the Qaeda

Operative Memo, although that opinion, unlike the Torture Memo, actually approved specific interrogation methods. Instead, the book emphasizes a memo from Yoo to the Defense Department dated March 14, 2003 (“Military Interrogation Memo”).48 Here, Yoo echoed both the Torture Memo’s analysis of law and the Qaeda Operative Memo’s justification of specific interrogation techniques. Goldsmith withdrew the March 14, 2003 opinion early in his tenure as director. Again, he did not do so because it approved specific interrogation methods, but because of its general legal analysis, which tracked the Torture Memo’s.49 In The Terror Presidency Goldsmith wrote:

45 See, e.g., GOLDSMITH, supra note 20, at 94–95 (suggesting that “law and lawyers were a big

part of the problem” that caused risk aversion to acting on intelligence prior to 9/11). See also discussion infra Conclusion (describing an example of Goldsmith’s views on the rule of law being misunderstood).

46 GOLDSMITH, supra note 20, at 150–51. 47 OPR Final Report, supra note 8, at 122. 48 Military Interrogation Memo, supra note 2. 49 GOLDSMITH, supra note 20, at 151–55.


[I]n a second August 1, 2002, opinion [the Qaeda Operative Memo]. . . OLC applied . . . [the Torture Memo’s] abstract analysis to approve particular and still-classified interrogation techniques. These separately and specifically approved techniques contained elaborate safeguards and were less worrisome than the abstract analysis in the public torture opinions themselves, which went far, far beyond what was necessary to support the precise techniques, and in effect gave interrogators a blank check. The same bifurcation occurred with the Defense Department: The . . . [Military Interrogation Memo] contained abstract and overbroad legal advice, but the actual techniques approved by the department were specific and contained elaborate safeguards.50

Thus, Goldsmith objected to OLC’s legal analyses, but not to its

conclusion that the actual interrogation methods were legal. Goldsmith and his views have been widely misunderstood. I argue

that this phenomenon can be traced to the historic framework. A kind of anti-torture chimera emerges from Goldsmith’s words and acts, something visible to those with a mental map of the framework in their heads. This effect, which is described in detail below, confounds the historic debate, confuses moral and ethical judgment, and obscures what actually went wrong at OLC.

E. The OPR Investigation

The Justice Department’s ethics investigators treat Goldsmith as an exemplary, almost canonical, figure. Both the OPR Report and the Margolis Opinion repeatedly cite him as authoritative.51 More important than these citations, the Justice Department adopted Goldsmith’s idea that the jurisprudence of the Torture Memo—rather than OLC’s approval of

50 Id. at 150–51. 51 E.g., Margolis Opinion, supra note 6, at 17–19 (citing a “lengthy, but instructive” exchange in

which Goldsmith gave OPR investigators a general overview of issues connected with “Lawyering for the President”); id. at 44 (citing Goldsmith’s criticisms of Yoo); id. at 46 (quoting Goldsmith’s comments in THE TERROR PRESIDENCY about relationships among the intelligence community, the Executive, and Congress); id. at 66 (noting that Goldsmith’s view that Yoo “certainly didn’t” think he was “violating the law”); OPR Final Report, supra note 8, at 17–18 (quoting at length from THE TERROR PRESIDENCY about the force of OLC opinions). See also Letter from Michael B. Mukasey, Att’y Gen. and Mark Filip, Deputy Att’y Gen., to H. Marshall Jarrett, Counsel. Office of Professional Responsibility, U.S. Dept. of Justice 14 (Jan. 19, 2009), available at http://judiciary.house.gov/ hearings/pdf/Mukasey-Filip090119.pdf (citing in the penultimate sentence of a lengthy discussion of OPR’s ethics investigation, what “Jack Goldsmith has written already”).


particular interrogation techniques as legal—was of significance. The OPR Report recommended professional discipline for Bybee and

Yoo without even “attempt[ing] to determine . . . whether the Bybee and Yoo memos arrived at a correct result.”52 As Margolis would later put it, OPR found Yoo and Bybee guilty of misconduct “not because they were wrong, but because they were not thorough.”53 Yet, Margolis himself seemed to treat the question of whether “they were wrong” as beside the point. Bybee’s and Yoo’s “most significant errors” were in the Torture Memo, Margolis wrote, yet the errors there “were not likely to cause prejudice” because the Qaeda Operative Memo, which had appeared at the same time, “approved specified techniques.”54 Thus, Margolis exonerated the Torture Memo’s authors partly because they had simultaneously approved waterboarding, stress positions, walling, and the other enhanced techniques. Margolis’ view of the Qaeda Operative Memo as exculpatory is akin to Goldsmith’s marveling at how OLC produced a pointed legal opinion—the Qaeda Operative Memo—on the same day as the overbroad Torture Memo.

Margolis followed Goldsmith’s lead in another respect. The Terror Presidency faults the Torture Memo most of all for damaging OLC’s reputation after it leaked.55 It seemed incongruous for Goldsmith to value OLC’s reputation so highly, particularly when so much else was at stake. Yet, Margolis echoes Goldsmith:

Even though the . . . [Torture Memo] was intended for a limited audience, Yoo and Bybee certainly could have foreseen that the memorandum would someday be exposed to a broader audience, and their failure to provide a more balanced analysis of the issues created doubts about the bona fides of their conclusions.56

Thus, the Torture Memo’s fault, Margolis implies, was its potential to

create public doubt. Yet how would we view an ethics investigation by another nation of officials who were alleged to have sanctioned torture—an investigation that dismissed the question of whether torture had occurred and focused instead the potential damage done to the reputation of the government agency that the officials worked for?

Part II introduces the historic framework and applies it to the

52 OPR Final Report, supra note 8, at 160. 53 Margolis Opinion, supra note 6, at 21. 54 Id. at 65. 55 See e.g., GOLDSMITH, supra note 20, at 158. 56 Margolis Opinion, supra note 6, at 68.


contemporary debate. Part III assesses OLC’s interrogation opinions in light of this framework. Part IV turns to Jack Goldsmith and examines his tenure at OLC. Part V applies the historic framework to Goldsmith’s views; it also examines the alternate conceptual framework—what I describe as a “chimera”—that emerges from his work.

II. THE HISTORIC FRAMEWORK

A. A Pattern of Debate

Writing 500 years ago, Beccaria said that torture arose from the idea that “pain should be the test of truth, as if truth resided in the muscles and fibres of a wretch in torture.”57 Since truth did not reside there, torture did not work; Beccaria argued that the victims would say anything to stop the torment.58 Even ancient Romans, though “for all their barbarity,” had limited its use, Beccaria observed.59 Torture was “worthy only of a cannibal.”60

On this view, torture reflects individual or national character, exposing torturers as barbarians or worse. As for the victims, what they say or do under torture bears no relation to who they really are or what they believe. Human identity and personality get reduced to nullities by torture: only “muscles and fibres” remain.61

A century later, Voltaire assessed not what victims of torture said, but what torture said about those who employed it. The practice traced back to highwaymen who were “in the habit of squeezing thumbs, burning the feet of those who refuse to tell them where they have put their money, and questioning them by means of other torments.”62 The next step in torture’s evolution was political, according to Voltaire:

[C]onquerors, having succeeded these thieves, found this invention of the greatest utility. They put it into practice

57 CESARE BECCARIA, OF CRIMES AND PUNISHMENTS [hereinafter OF CRIMES AND

PUNISHMENTS], available at http://www.constitution.org/cb/crim_pun16.htm. A recent edition of Beccaria includes a slightly different translation: “physical suffering comes to be the crucible in which truth is assayed, as if such a test could be carried out in the sufferer’s muscles and sinews.” CESARE BECCARIA, Of Crimes and Punishments, in CRIMES AND PUNISHMENTS AND OTHER WRITINGS 1, 39 (Richard Bellamy ed., Richard Davies et. al. trans., Cambridge University Press 1995) [hereinafter BECCARIA].

58 BECCARIA, supra note 57, at 41. 59 Id. at 40–41. 60 Id. at 40. 61OF CRIMES AND PUNISHMENTS, supra, note 57, at *1. 62 VOLTAIRE, PHILOSOPHICAL DICTIONARY 394 (Theodore Besterman ed., 1972) (1764).


when they suspected that some vile plot was being hatched against them, as, for instance, that of being free . . . .63

Thus, torture marks its practitioners as criminal and uncivilized,64

revealing more about the torturers than anything else. Torture victims may possibly recover, but torturers remain barbaric and criminal.

Critics of harsh interrogation in the War on Terror echo the classic criticisms.65 They too make claims about the immorality and inhumanity of torture. Like Beccaria and Voltaire, they emphasize torture’s baleful effects on its perpetrators.

Thus, Lord Hoffmann, argues that torture “corrupts and degrades the state which uses it and the legal system which accepts it.”66 Sanford Levinson notes the “extraordinary importance” of torture policies “to

63 Id. Cf. “[T]he torturer has become—like the pirate and slave trader before him—hostis humani generis, an enemy of all mankind.” Waldron, supra note 10, at 1719 n.170 (quoting Filartiga v. Pena-Irala, 630 F.2d 876, 890 (2d Cir. 1980)).

64 Voltaire cites the example of Russians, who had been “regarded as barbarians” in 1700, but where torture was abolished in mid 18th century. VOLTAIRE, supra note 62, at 396. Abolition of torture ranked second only to “universal toleration” as a mark of Russian enlightenment, in Voltaire’s view. Id

65 The same figures of speech are sometimes used. For example, Jeremy Waldron described the “spirit” of a ban on torture in the following way:

Law is not brutal . . . . Law is not savage. Law does not rule through abject

fear and terror, or by breaking the will of those whom it confronts . . . . People . . . will not be herded like cattle or broken like horses; they will not be beaten like dumb animals or treated as bodies to be manipulated . . . .[Legal] force and coercion do not work by reducing [persons] to a quivering mass of bestial, desperate terror.

Waldron, supra note 10, at 1726–27 (internal quotations and references omitted). A version of the same argument was made in connection with the drafting of the European Convention on Human Rights, when a delegate condemned torture as “retrogression into barbarism” and proposed a declaration that “all forms of physical torture . . . are inconsistent with civilized society.” Waldron, supra note 10, at 1710 (quoting 2d Council of Eur., Collected Edition of the “Travaux Préparatoires” of the European Convention on Human Rights, 36–38, 40 (1975)). Making a related point, Kutz argues that Bush administration interrogation policies “can be seen as rejecting two of the deepest legacies of the Enlightenment: the inviolability of the individual and the subordination of power to principles of right.” Kutz, supra note 3, at 238.

66 A v. Secretary of State for the Home Department [2005] UKHL 71, 2 A.C. [82] quoted in Baroness Hale of Richmond, Human Rights in the Age of Terrorism: The Democratic Dialogue in Action, 39 GEO. J. INT’L L. 383, 400–01 (2008). Lord Hoffmann also cites Beccaria and Voltaire:

When judicial torture was routine all over Europe, its rejection by the

common law was a source of national pride and the admiration of enlightened foreign writers such as Voltaire and Beccaria. In our own century, many people in the United States, heirs to that common law tradition, have felt their country dishonoured by its use of torture outside the jurisdiction and its practice of extra-legal “rendition” of suspects to countries where they would be tortured.

Id.

http://web2.westlaw.com/find/default.wl?tf=-1&rs=WLW9.11&referencepositiontype=S&serialnum=1980125293&fn=_top&sv=Split&referenceposition=890&pbc=9854064A&tc=-1&ordoc=0305997214&findtype=Y&db=350&vr=2.0&rp=%2ffind%2fdefault.wl&mt=208

http://web2.westlaw.com/find/default.wl?tf=-1&rs=WLW9.11&referencepositiontype=S&serialnum=1980125293&fn=_top&sv=Split&referenceposition=890&pbc=9854064A&tc=-1&ordoc=0305997214&findtype=Y&db=350&vr=2.0&rp=%2ffind%2fdefault.wl&mt=208


defining who we are as a people.”67 And Christopher Kutz observes that “debates over the bounds of interrogation . . . go to the heart of national values and identity”68 since the “principles” that torture violates “have become the marks by which we know our moral identities as both persons and nations.”69 Meanwhile, it has become almost a commonplace that torture undermines the “rule of law,”70 a criticism President Obama articulated in these words: “[W]e’ve been the nation that has shut down torture chambers and replaced tyranny with the rule of law. That is who we are.”71

A related argument is that torture degrades the United States by reducing it to the level of its adversaries. If the United States “answer[s] . . . terrorism with torture,” writes Stephen Holmes, it “will certainly lose the moral dimension of its war on terror . . . .”72 Sanford Levinson makes a related point, along with a comparison to ancient Rome somewhat different from Voltaire’s:

67 Sanford Levinson, Contemplating Torture: An Introduction, in TORTURE: A COLLECTION 23, 38 (Sanford Levinson ed., 2004) [hereinafter TORTURE: A COLLECTION]. It is easy to multiply examples. See, e.g., Seth F. Kreimer, Too Close to the Rack and the Screw: Constitutional Constraints on Torture in the War on Terror, 6 U. PA. J. CONST. L. 278, 300, 310 (2003) (arguing that torture implicates “national identity” and “the nature of our society”); David Horsey, On Torture: Are We Roman or American?, DAVID HORSEY’S, DRAWING POWER, http://blog.seattlepi.com/davidhorsey/ archives/168850.asp (May 15, 2009, 11:02 PM) (likening “our descent into torture” to brutal practices in ancient Rome, so that “we, today, have become too Roman”); Waldron, supra note 10, at 1687 (describing torture as “utterly repugnant to the spirit of our law”). President Obama has sounded similar themes discussing his administration’s approach to combating terrorism. In his Inaugural Address President Obama said “Our founding fathers, faced with perils that we can scarcely imagine, drafted a charter to assure the rule of law and the rights of man, a charter expanded by the blood of generations. Those ideals still light the world, and we will not give them up for expedience’s sake.” President Barack Obama, All This We Will Do, Inaugural Address (Jan. 20, 2009), reprinted in N.Y. TIMES, Jan. 21, 2009, at A2.

68 Kutz, supra note 3, at 237. 69 Id. at 275. 70 See, e.g., Jane Mayer, The Battle for a Country’s Soul, N.Y. REV. BOOKS (July 15, 2008),

http://www.nybooks.com/articles/archives/2008/aug/14/the-battle-for-a-countrys-soul/ (noting “tragically destructive long-term consequences both for the rule of law and America’s interests in the world”); Major David J.R. Frakt, Closing Argument at Guantanamo: The Torture of Mohammed Jawad, 22 HARV. HUM. RTS. J. 1, 8 (2009) (expressing hope that the author’s legal arguments “might . . . help nudge the country back in the direction of restoring the primacy of the rule of law”); Editorial, A Crack in the Wall of Secrecy, N.Y. TIMES, Apr. 26, 2009 (describing the release of interrogation memos as “an essential step toward re-establishing the rule of law”).

71 Steve Coll, Op-Ed., Threats, NEW YORKER, Jan. 18, 2010, at 19 (quoting President Barack Obama, Address to the National Archives (May 2010)). Coll added that President Obama’s “counterterrorism strategy” is “premised upon a forward defense and the durability of American constitutional values.” Id.

72 Stephen Holmes, Is Defiance of Law a Proof of Success? Magical Thinking in the War on Terror, in THE TORTURE DEBATE IN AMERICA 118, 133 (Karen J. Greenberg ed., 2006) [hereinafter THE TORTURE DEBATE IN AMERICA].


If the United States is widely believed to accept torture as a proper means of fighting the war against terrorism, then why should any other country refrain? The United States is . . . the “new Rome,” the giant colossus bestriding the world and claiming . . . to speak in behalf of good against evil.73

Opponents of an absolute ban on torture concede the immorality of the

practice, or at least consider its use problematic. But they also consider torture necessary on occasion—a lesser evil—such as when torture will reveal the location of a “ticking bomb.”74 More generally, they believe it is sometimes necessary to “fight[] fire with fire,”75 the presumption being that torture works.

Richard Posner in particular argues that history has “falsified” the claim that “recourse to torture so degrades a society that it should be forsworn even if the death of many innocents” results.76 Posner cites the examples of “France (in Algeria), the United Kingdom (in its struggle with the Irish Republican Army), and Israel (in combating the intifada)”—each of which “used torture to extract information” but did not, according to Posner, lapse into total barbarism as a result.77

B. Beccaria’s Trope

Beccaria opened the historic debate with an implicit contrast between abstract entities—personhood, nationhood, civilization—and physical properties, such as nerves and fibers.78 Remarkably, this simple trope established the subsequent framework of debate. Anti-torture arguments rest on the existence of such abstract entities and on the damage that torture does to them. Opposing arguments slight these abstractions or ignore them entirely.

Although nations appear to be abstract entities, they feature prominently in arguments for harsh interrogation. A nation’s survival may even be the deciding consideration. In such arguments, however, national

73 Levinson, supra note 67, at 38. 74 Id. at 30–33 (discussing arguments for a non-absolute ban on torture); Kate Kovarovic, Our

“Jack Bauer” Culture: Eliminating the Ticking Time Bomb Exception to Torture, 22 FLORIDA J. INT. L. 251, 254–57 (2010) (canvassing “ticking time bomb” arguments).

75 Richard A. Posner, Torture, Terrorism, and Interrogation, in TORTURE: A COLLECTION, supra note 67, at 294.

76 Id. 77 Id. Posner does not attempt to determine, however, whether torture damaged those nations’

social, moral and legal fabrics short of reducing them to total barbarism. 78 See generally BECCARIA, supra note 57.


survival typically means physical survival. Nations are treated as collections of physical objects—like Beccaria’s image of persons reduced to their “muscles and fibres”—rather than as abstract entities with an internal life and unique characters, values, or souls.79

Institutions smaller than nations receive similar treatment. When military lawyers objected to harsh interrogation methods during the Bush administration, for example, they cited the institutional character and values of the military services. “[C]oercive interrogations,” wrote the General Counsel of the Navy, threatened to “profoundly alter[] . . . [the U.S. military’s] core values and character[.]”80 In the same vein, the Judge Advocate General of the Navy, Admiral Michael F. Lohr, wondered whether “the American people [will] find [that] we have missed the forest for the trees by condoning practices that . . . are inconsistent with our most fundamental values[.]”81

One can find departures from the argumentative pattern, but the departures themselves often reflect the framework. Consider national character. It is not argued today that torture enhances or reflects national character. Instead, national character is generally ignored in pro-interrogation arguments.

One exception to this trend, Richard Posner, was just noted. Not ruling out torture, Posner did consider national character. However, he disposed of it in a cursory way. Posner argued for precise cost-benefit analyses of torture’s possible benefits.82 Yet, his assessment of torture’s effects on national character was neither precise nor measured; instead, it was a binary, all or nothing argument. Posner observed that France, England and Israel, after employing torture, did not relapse into complete barbarism,83 but he never attempted to assess damage to national character short of total moral and legal collapse. Turning around this all or nothing approach, one

79 See supra note 57 and accompanying text. Goldsmith’s book provides an example. See infra notes 265–66 and accompanying text.

80 Memorandum from Alberto J. Mora, Gen. Counsel of the Dep’t of the Navy, on Statement for the Record: Office of General Counsel Involvement in Interrogation Issues to Inspector Gen., Dep’t of the Navy 11 (July 7, 2004), available at http:// www.newyorker.com/images/pdfs/moramemo.pdf [hereinafter Mora Memorandum]. See also Levi, supra note 35, at 1458. Mora’s role is discussed in Chapter 9 of THE DARK SIDE. See generally MAYER, supra note 44, at 213–37.

81 Memorandum from Michael F. Lohr, Rear Admiral, Judge Advocate Gen., U.S. Navy, on Working Group Recommendations Relating to Interrogation of Detainees to Gen. Counsel of the Air Force (Feb. 6, 2003), in THE TORTURE DEBATE IN AMERICA, supra note 72, at 382 [hereinafter Lohr Memo]. Admiral Lohr’s formal recommendation was “[a]t a minimum” to limit the techniques to the “very narrow set of circumstances” presented by detention at the Guantanamo Bay prison. Id., paras. 2, 4. In an interesting ambiguity, Admiral Lohr left it unclear whether the “fundamental values” he was citing defined the Navy, the United States, or both.

82 See Posner, supra note 75, at 297. 83 Id. at 294.


might argue that torture is warranted only if it will save the United States from total destruction.84 Since only a complete reversion to barbarism counted against torture, it would seem that only the certain avoidance of complete national destruction should count in its favor.

The difficulty traces back to the argumentative framework. The abstractions in anti-torture arguments—humanity, civilization, national and individual identity, rule of law—are ill-adapted to cost-benefit analyses. Instead, they operate like argumentative trumps. If something is inhumane, uncivilized or destructive of personal or national identity, it simply should not be done. Thus, when Posner tried to take those anti-torture concepts into account, he did not engage in cost-benefit calculations. Posner stretched the historic pattern by considering national character—rather than completely ignoring it—but he still did not escape the framework completely.

Anti-torture writers face the mirror image of this problem. Populated with argumentative trumps, anti-torture arguments often shy away from particulars, comparative assessment, and measurement. Thus, anti-torture writers sometimes assume that torture is completely ineffective for acquiring information, as Beccaria had suggested, although that may or may not be true.

The conceptual problems for anti-torture writers may be less serious, however. Since anti-torture concepts operate as trumps, torture should be banned even if it is effective. Voltaire, for example, assumed that torture worked well for highway robbers pursuing loot, but he still condemned it.85 Jeremy Waldron, an anti-torture writer, has dealt fairly easily with the possibility of torture causing only minimal damage to the rule of law. Waldron predicted devastating, pervasive effects on law if torture was allowed, but he also said that if those effects did not materialize, he might change his mind about torture. “Presented with solid evidence,” he wrote, “that a legal system that permitted torture was nevertheless able to maintain the rest of the adjacent law about nonbrutality intact over the long or medium term, I would have to abandon my concern about the systemic effects of messing with [torture] provisions.”86

Thus, one side favors abstract entities and argumentative trumps, while the other focuses on material things, specific measurements, and

84 Waldron considers bans on torture the “archetype” of a deeper, anti-brutality principle in law, but he observes that other rules against brutality would “not . . . necessarily unravel[] the instant we diminish the force of the archetype.” Waldron, supra note 10, at 1748. Posner assumes, on the other hand, that no damage is done to the legal system unless the damage is total—a notable departure from his general view that phenomena relevant to law are measurable and that law should measure them. Posner, supra note 75, at 291–98.

85 Voltaire, supra note 62, at 394–95. 86 Waldron, supra note 10, at 1735.


balancing.87 Given this dichotomy, it is ironic that the more intent the focus on specific tortures or individual cases, the more inclined one becomes to think in terms of abstractions like humanity or personhood.88

C. Apparent Exceptions

The persistence of these argumentative patterns is underscored by some apparent modern exceptions—exceptions that on closer analysis reflect the historic pattern after all.

1. Torture Warrants

Consider, for instance, Alan Dershowitz’s proposed system of “torture warrants”—that would authorize official torture—and which he believed would enhance the rule of law.89 Before an official tortures someone, Dershowitz argued, a warrant should be obtained.90 He summarized his position as follows:

I am opposed to torture as a normative matter, but I

know it is taking place today and believe that it would certainly be employed if we ever experienced an imminent threat of mass casualty biological, chemical, or nuclear terrorism . . . .[I]f torture is being or will be practiced, is it worse to close our eyes to it and tolerate its use by low-level law enforcement officials without accountability, or instead to bring it to the surface by requiring that a warrant of some kind be required . . .91

It is striking to encounter appeals to the “rule of law” and

“accountability” in an argument for torture warrants. The argument

87 Compare Posner, supra note 75, at 293 (describing his cost effective balancing approach), with Waldron, supra note 10, at 1701 ( “There are some scales one really should not be on, and with respect to which one really does not have a legitimate interest in knowing precisely how far along the scale one is permitted to go.”).

88 For example, when Goldsmith actually saw an imprisoned detainee he momentarily questioned Bush administration policies. GOLDSMITH, supra note 20, at 102. Thus, the very detail favored by pro-torture arguments may lead to an anti-torture conclusion. (Not always, however, since one might also focus on the victims of attacks that arguably could have prevented by torture.).

89 Alan Dershowitz, Tortured Reasoning, in TORTURE: A COLLECTION, supra note 67, at 257. 90 ALAN M. DERSHOWITZ, WHY TERRORISM WORKS: UNDERSTANDING THE THREAT,

RESPONDING TO THE CHALLENGE 158 (2002); Dershowitz, supra note 89, at 257, 273 (noting, and apparently disapproving, the French prosecution of a general “not for what he had done to the Algerians . . . [but] for revealing what he had done and seeking to justify it”).

91 Dershowitz, supra note 89, at 257. See also id. at 273 (“No legal system operating under the rule of law should ever tolerate an “off-the-books approach.”); id. at 275–76 (“At bottom, my argument is not in favor of torture of any sort. It is against all forms of torture without accountability.”).


centers on an abstraction, the rule of law, yet it contemplates torture.92 Upon closer examination, however, this proposal does not substantially

depart from the argumentative pattern. Certainly, there is no claim that torture ennobles the nation or that the rule of law requires its use. Strictly speaking, Dershowitz does not argue that the United States should use torture; he only assumes that the United States will do so. Having shifted issues in this way, Dershowitz’s proposal really does not belong to the “torture versus no torture” debate at all.93 Consistent with the pattern, however, Dershowitz implicitly rejects conceptions of the rule of law as a highly integrated structure, the very existence of which is threatened by torture.94

Dershowitz’s proposal was controversial, and a debate ensued about whether warrants would limit or encourage the practice. Viewed in light of the historic pattern, the proposal would seem pointless or self-contradictory. Historically, those opposed to torture believe that any use of it undermines the rule of law; hence, the rule of law could not possibly be part of an argument for torture warrants. If there was a warrant proceeding, on the classical anti-torture view, all applications should be denied. Dershowitz had positioned his proposal just outside the historic mold, and the resulting debate was about whether his proposal came so close that it should be subsumed within the historic pattern. By skirting so close to the line, Dershowitz provided further evidence that the line exists.

2. John T. Parry

John T. Parry stretches the historic pattern in a different direction. Like Dershowitz, Parry believes the United States employs torture and will continue doing so. Nor would Parry categorically rule it out, even though he believes that torture is used too frequently. Unlike Dershowitz,

92 Broaching his proposal in the Los Angeles Times, Dershowitz wrote, “Democracy requires accountability and transparency, especially when extraordinary steps are taken. Most important, it requires compliance with the rule of law. And such compliance is impossible when an extraordinary technique, such as torture, operates outside of the law.” Alan M. Dershowitz, Is There a Torturous Road to Justice?, L.A. Times, Nov. 8, 2001, at B19, available at http://articles.latimes.com/print/2001/ nov/08/local/me-1494.

93 Dershowitz notes that his proposal is not part of “the old, abstract Benthamite debate over whether torture can ever be justified.” Dershowitz, supra note 89, at 266. He also suggests that his proposal would “perhaps reduc[e]” the “frequency and severity” of torture’s use. Id. at 267.

94 Early in his career, Dershowitz documented extensive systems of confinement and coercion that ignored conventional rule of law principles—notably, civil commitment and the powers traditionally exercised by Justices of the Peace. See generally Alan M. Dershowitz, Preventive Confinement: A Suggested Framework for Constitutional Analysis, 51 TEX. L. REV. 1277 (1973). Dershowitz argued for extending the rule of law and democratic accountability to those areas, a position with obvious connections to the later torture warrant proposal. Id. These earlier investigations may have led Dershowitz to conclude that the rule of law is neither as homogeneous nor as vulnerable a structure as is sometimes believed.


however, Parry “favor[s] an ex post approach to the problem of interrogational torture” and believes that “after-the-fact application of doctrines such as the necessity defense provides the best way to address the very rare instances in which torture could be justified.”95

Like classic anti-torture writers, Parry accepts the idea of a highly integrated “legal and political culture” that is relevant to torture.96 The next step in standard anti-torture arguments would cite the damage done to that culture by torture. Here, however, Parry departs from the pattern. He finds American culture and law hospitable, rather than antagonistic, to torture.97 To support this conclusion, Parry cites a history of torture by the United States in foreign countries;98 harsh and coercive American prison practices; the continuation of brutal interrogations and other police methods, despite judicial decisions banning them; and consistently narrow interpretations of treaties and international agreements regarding torture and other human rights by a succession of American administrations.99 Yet, Parry is certainly not suggesting that the United States should nurture its culture by using more torture.

In another departure from standard arguments, Parry does not consider the law a promising vehicle for limiting torture. Instead, he applauds political anti-torture activists—although not their categorical opposition to torture—because they have “done a great deal of important work to publicize and restrict torture and obtain remedies or help for victims of torture.”100 “[P]eople who really want to combat torture,” he writes, “should engage in more direct responses—for example, in political action—and pay less attention to law. Law, after all, failed to control, and indeed was interpreted to license, the systematic abuse of people detained in the ‘war on terror.’”101

While Parry believes some good can come from his “after the fact”

95 But compare John T. Parry, Torture Warrants and the Rule of Law, 71 ALB. L. REV. 885, 886

(2008) (limiting the use of such defenses), with Torture Memo, supra note 1, at 207–09 (describing a robust necessity defense available to interrogators accused of violating the anti-torture statute).

96 John T. Parry, Torture Nation, Torture Law, 97 GEO. L.J. 1001, 1003 (2009). 97 “[T]orture may be compatible with American values in practice and with the legal system . . .

[that] serve[s] those values,” Parry observed. Id. He concluded the “appropriate descriptive narrative for torture and abuse in the war on terror is less one of disjuncture and more one of continuity with the rule of law as a domestic practice.” Id. at 1056.

98 Id. at 1005–16. 99 Id. at 1032–52. But cf. Jerome H. Skolnick, American Interrogation: From Torture to

Trickery, in TORTURE: A COLLECTION, supra note 67, at 105, 122 (arguing that judicial decisions “have virtually eliminated interrogatory torture in the United States,” at least when the police are seeking admissible evidence).

100 Parry, supra note 95, at 890. 101 Id.


proposal for appraising torture, he has limited hope for any legal measure designed to curb the practice. Finding harsh measures deeply ingrained in the culture, and considering “culture” a meaningful, persistent entity, Parry could hardly suppose that a few legal reforms would fundamentally change things.102 Regarding culture as extraordinarily powerful, Parry perhaps considers it more important than any logical or legal category—a position that distinguishes him from many anti-torture authors.

Like Dershowitz, Parry falls outside the historic pattern, without actually running afoul of it. His harsh view of contemporary American legal culture hardly puts him at odds with Beccaria or Voltaire; were they alive, they might agree with him. But Beccaria and Voltaire would go on to appeal to civilized norms, which exist beyond the norms of national culture. Parry himself seems too much the realist to do that; he does not appeal to those norms because he seemingly does not recognize them as real. The result is that he wants less torture, but it remains unclear why. Yet Parry does not really break the pattern; he never argues for more torture based on abstractions or cultural norms.

The historic framework is like a force-field. Entirely outside the field lie arguments that would cite abstractions to justify torture—arguments that the historic framework has largely eliminated from the current debate. Other arguments, however, like Parry’s and Posner’s, show the influence of the framework without being completely under its control.103

III. OLC AND OPR

The OPR Final Report cited ethical flaws in the Torture Memo that, according to investigators, warranted ethical censure.104 The Justice Department’s reviewer, Margolis, then identified flaws in the OPR investigators’ own ethical arguments.105 I argue, in turn, the investigators and Margolis both erred. All this suggests the prospect of never ending arguments about mistaken attributions of error. As noted in the Introduction, it is a reason to look to the framework for help.

Despite that, the inference from legal error to ethical failing remains a viable one. Indeed, the very narrowest legal arguments—for example,

102 Parry, supra note 95, at 893 (“Torture, I am suggesting, is part of the fabric of modern legal

and political life. It also follows that, if we believe we live in a society committed to and founded upon the rule of law, then the rule of law is no certain bulwark against torture, and faith that it could be is seriously misguided.”).

103 Another example is an argument for less harsh interrogation based on calculations, weighing, or cost-benefit analyses. A claim that torture “does not work well enough to justify its use” would fall into this category.

104 OPR Final Report, supra note 8, at 254, 256–57. 105 Margolis Opinion, supra note 6, at 2.


those purporting to rely on dictionary definitions of “torture”—would show no influence from the framework at all. In the case of these narrow legal arguments—for instance, a badly botched dictionary definition of “severe pain and suffering”—that inference from error will often prove more revealing than anything else.

A) Some Hypothetical Arguments

Consider, then, six possible arguments for the legality of harsh interrogations:

1) Article II of the Constitution explicitly gives the

President a power to torture; 2) The anti-torture statute explicitly permits

waterboarding, sleep deprivation, and extended stress positions;

3) Properly read, Article II makes the President the sole judge of the extent of executive power;

4) Every constitutional or statutory right is subject to a balancing test: no rights—including the right to be free from torture or cruel and degrading treatment—is absolute;

5) The anti-torture statute, properly interpreted, requires prolonged physical suffering before an interrogation technique can qualify as torture;

6) Neither waterboarding, sleep deprivation, nor extended stress positions qualifies as torture under the statute because those techniques—alone and in combination—do not inflict severe physical or mental pain or suffering.

Arguments #1 and #2 falsely claim that the Constitution or the anti-

torture statute includes an explicit provision. The framework sheds no light on these arguments; it simply has no bearing on whether a law includes specific language. On the other hand, an inference based on blatant legal error in these cases would be strong.

Arguments #3 and #4 also lend themselves to the standard approach, but not as well. Both arguments are overbroad and are seemingly contradicted by case law. The political question doctrine on occasion gives presidents the authority claimed by argument #3, but in general, presidential actions are judicially reviewable. Similarly, some constitutional rights are subject to balancing, but not all. Yet, these errors are less stark than false claims about language in a statute; someone could adopt these positions based on what many would consider purely legal grounds. Thus, Margolis noted that John Yoo had subscribed to expansive


views of executive power as a law professor, and he downgraded the seriousness of Yoo’s alleged errors in the Torture Memo as a result.106

The framework sheds a different and somewhat attenuated light on these arguments. To the extent that argument #3 rests on the idea that decisions confronting a president always require particularized assessments and risk benefit calculations—and that the decisions therefore can never be subsumed under abstract or law-like principles—it echoes a tenet of the classic pro-torture position. The same holds for argument #4, which comes close to claiming that no fundamental human rights exist in law. These affinities with the framework may be interesting, but they also seem less than compelling.

Argument #5—which asserts that physical suffering qualifies as torture only if the suffering is prolonged—interprets the anti-torture statute. Unlike arguments #1 and #2, it does not merely assert that certain words appear in the law. Yet, as was true of those earlier arguments, the historic framework seemingly sheds little light. Unlike arguments #1 and #2, however, this argument is not merely hypothetical; it appears in the interrogation opinions, and is discussed below.

Argument #6 also appears in the OLC opinions. “Severe physical or mental pain or suffering” is part of the statutory definition of torture. In part, then, argument #6 rests on an interpretation of the relevant statutory terms—a matter that, as just noted, the framework rarely helps to clarify. Here, however, OLC’s approach to that question will turn out to depend on presumptions about persons and the world at large that echo classic pro-torture positions. Here, the analyses regarding error and the framework overlap.

I now turn to OLC’s arguments, and OPR’s ethical judgments.

B. OLC’s Arguments and OPR’s Judgments

1. “Physical Suffering”

To constitute torture, the anti-torture statute requires either physical or mental pain or suffering.107 The Torture Memo interpreted physical “pain or suffering” as a single undifferentiated entity, so that no category of “physical suffering” existed apart from the category of “physical pain.”108 Recognizing that “[o]ne might argue” the other way, the Memo supported its interpretation by making three points.109

First, it looked to the statute’s explicit definition of “severe mental

106 Margolis Opinion, supra note 6 at 67. See also supra text accompanying notes 13–16. 107 §§ 2340–2340A. See also supra note 1. 108 Torture Memo, supra note 1, at 176–77 n.3. 109 Id.


pain or suffering” for guidance about the meaning of “physical pain or suffering.”110 Treating “mental pain or suffering” as a single entity, the statute provides that mental pain or suffering does not exist absent “prolonged mental harm.”111 Drawing a parallel between the physical and the mental, Yoo argued in the Torture Memo that the drafters of the statute must have considered physical “pain or suffering” a “single concept” too.112 Second, Yoo cited dictionary definitions defining pain in terms of suffering, and vice versa. Third, Yoo argued that “even if we were to read the infliction of severe physical suffering as distinct from severe physical pain, it is difficult to conceive of such suffering that would not involve severe physical pain.”113

However, waterboarding, which produces a sense of death by drowning and suffocation, involves exactly what Yoo described as “difficult to conceive”—severe physical suffering without actual pain. The Techniques Memo, for example, noted that “the waterboard technique is not physically painful” although it produces physical “distress.”114 Moreover, if “pain or suffering” constitutes one undifferentiated entity, as Yoo argued, then pure “suffering” should qualify as “pain or suffering” as much as pure “pain” does. (The same conclusion follows if “pain” and “suffering” have the same meaning.) Thus, if the sensation of drowning qualifies as “severe physical suffering”—and it surely does—waterboarding should qualify as torture.

Abandoning the analysis in the Torture Memo, the Techniques Memo, which replaced it, recognized a separate statutory category of “physical suffering.” But it also interpreted the statute to prohibit severe suffering only if the suffering was both intense and prolonged.115 Applying that interpretation, the memo concluded that

the physical distress caused by the waterboard would not be expected to have the duration required to amount to severe physical suffering [because] [a]pplications [of the waterboard] are strictly limited to at most 40 seconds, and a total of at most 12 minutes in any 24 hour period, and

110 Id. at 177. 111 Id. See also § 2340(2) (“‘severe mental pain or suffering’ means the prolonged mental harm

caused by or resulting from . . .”). 112 Id. at 177 n.3. 113 Id. 114 Techniques Memo, supra note 2, at 70–71. 115 Id. at 51 (“We therefore believe that ‘severe physical suffering’ under the statute means a

state or condition of physical distress, misery affliction, or torment, usually involving physical pain, that is both extreme in intensity and significantly protracted in duration or persistent over time.”).


use of the technique is limited to at most five days during the 30-day period we consider.116

This requirement of prolonged physical suffering first appeared in the

Legal Standards Memo of December 30, 2004, written by Daniel Levin.117 Levin conceded that “physical torture is not limited to ‘severe physical pain.’”118 He immediately added, however, that the definition of “physical suffering” could not be “so broad as to negate the limitations on the other categories of torture in the statute.”119 By that statement, Levin apparently meant that because mental “pain and suffering” had to involve prolonged harm in order to constitute torture, physical suffering had to be prolonged in order to qualify as torture too.

The argument seems clearly wrong. A more obvious inference from the statute’s requiring “prolonged mental harm” in connection with mental pain or suffering, but not with physical pain or suffering, is that no duration requirement exists for physical suffering. Moreover, if Levin’s argument were correct, it would apply equally to physical pain—meaning that interrogators could pull out fingernails, administer powerful electric shocks or otherwise inflict “severe pain” during forty second intervals, for a total of twelve minutes in a twenty-four hour period, just as in the case of waterboarding. Yet nothing suggests anyone believed such a thing. Nor would it make sense to define “physical pain” as “prolonged physical harm”—the way that the statute defined “mental pain or suffering” in terms of prolonged mental harm—since physical pain and physical harm are so obviously different.

Although not considered a flaw by OPR,120 and often overlooked by commentators,121 OLC’s interpretations of “physical suffering” are among its most questionable. The Torture Memo and the Legal Standards Memo read the statutory language in different ways, but both readings are contrived to avoid the obvious conclusion that severe physical suffering, regardless of duration, constituted torture. This evasion was clearly necessary to uphold waterboarding, if not other interrogation techniques.

116 Id. at 71. 117 Legal Standards Memo, supra note 2, at 12. 118 Id. at 10. 119 Id. at 12. 120 OPR did note that the definition of “physical suffering” was seen as critical by some Justice

Department lawyers. See OPR Final Report, supra note 8, at 142 n.113 (noting the concerns of James Comey and others).

121 But see MAYER, supra note 44, at 306 (depicting the Legal Standards Memo as a “step toward restoring America’s laws and values,” but noting an “oddly contradictory” definition of torture “that seemed specifically written to legalize waterboarding so long as the severe pain [sic] it produced was not of ‘extended duration.’”).


Here, the 2004 and later OLC opinions, which supposedly improve on Yoo’s legal analysis, seem even more clearly wrong. However that may be, the “blatant error” analysis affords a strong basis for moral judgment.

2. The Article 16 Memo

A less categorical version of argument #4 (regarding balancing tests) appears in the Article 16 Memo, written by Stephen Bradbury in 2005.122 The question presented in this Memo was whether CIA interrogations constituted cruel, inhuman or degrading treatment.123 The answer, according to Bradbury, depended on whether the harms associated with interrogation were justifiable in light of the government’s interests.124

Bradbury found that they were justifiable. Even though a “strong tradition against” the techniques was evident in the practice of American criminal justice, the history of American military interrogation, and the condemnations issued by the State Department when other nations used similar methods,125 Bradbury determined that the government’s interests prevailed. This was so because of the seriousness of the threat, the need to protect “the United States and its interests,” and the difficulty of collecting intelligence in other ways.126 Bradbury said nothing about the seemingly absolute quality of a ban on cruel, inhuman or degrading treatment—a ban that, like the prohibition on torture itself, seems almost empty if a balancing test applies.

Bradbury’s analysis produced an anomaly. The massive weight of the government interests that he identified might justify all interrogation methods under exigent circumstances, including torture. Yet in light of the government’s interests, those methods would not qualify as cruel, inhuman or degrading treatment. In the Torture Memo, Yoo demonstrated convincingly that torture is generally understood to be significantly harsher than conduct that is cruel, inhuman and degrading.127 “Torture,” he wrote, “is a step far-removed from other cruel, inhuman or degrading treatment. . .

122 Article 16 Memo, supra note 2, at 124–26. See also Military Interrogation Memo, supra note

2, at 47 (also employing a totality of the circumstances test). 123 The ban on such treatment was contained in Article 16 of the United Nations Convention

Against Torture and Other Cruel Inhuman or Degrading Treatment Or Punishment art. 16, 26, Dec. 10, 1984, 1465 U.N.T.S. 116. Because of geographical limitations, OLC concluded that the Convention did not apply to CIA interrogation programs. Article 16 Memo, supra note 2, at 100–28. However, the CIA requested advice regarding the consistency of its interrogation with the substantive standards of the Convention, apart from any geographical limitations. Id. at 100, 123.

124 Article 16 Memo, supra note 2, at 126. 125 Id. at 129–36. 126 Id. at 136. 127 Torture Memo, supra note 1, at 184–91.


.”128 On that view, all torture is “cruel, inhuman or degrading”—but not all cruel, inhuman or degrading treatment rises to the level of torture. Bradbury, however, either eliminated the distinction between torture and cruel, inhuman and degrading treatment or else turned it on its head with a standard that suggested torture need not be cruel or inhuman.129

Bradbury even saw this difficulty. He wrote:

We do not conclude that any conduct, no matter how extreme, could be justified by a sufficiently weighty government interest . . . Rather, our inquiry is limited to the program under consideration, in which the techniques do not amount to torture considered independently or in combination.130

But this passage is almost a reductio ad absurdum of the admonition to write narrow opinions. Here, Bradbury effectively says that his opinion is too narrow to be wrong—even though it espouses a principle that may be questionable—since it is “limited to the program under consideration.”131

The discussion of cruel, inhuman and degrading treatment suffered from another flaw: its treatment of international precedent. In Ireland v. the United Kingdom, the European Court of Human Rights had found that wall standing, hooding, loud noises, and deprivation of sleep and food and drink amounted to inhuman and degrading treatment under the European Convention, but did not constitute torture.132 The methods resembled those used by the CIA, although waterboarding—the most suspect of the CIA techniques—was not employed.

The Torture Memo discussed Ireland and other international decisions at length, implying that they were accurate guides to the meaning of torture in American law.133 The international cases “make clear,” Yoo wrote,

that while many of these [interrogation] techniques may amount to cruel, inhuman or degrading treatment, they do not produce pain or suffering of the necessary intensity to

128 Id. at 191. 129 Article 16 Memo, supra note 2, at 123–28. 130 Id. at 128. 131 Id. 132 Torture Memo, supra note 1, at 197 (citing Ireland v. United Kingdom, 1976 Y.B. Eur. Conv.

on Hum. Rts. 512, 748 (Eur. Comm’n of Hum. Rts.)). 133 See id. at 196. Although Yoo believed that international opinions “are in no way binding

authority upon the United States,” he nonetheless observed that “[i]nternational decisions can prove of some value” in this instance. Id.


meet the definition of torture. From these decisions, we conclude that there is a wide range of such techniques that will not rise to the level of torture.134

That, in turn, supported the Torture Memo’s conclusion that “torture” was limited to “extreme conduct, resulting in pain that is of an intensity often accompanying serious physical injury.”135

The OPR investigation faulted the Torture Memo’s treatment of Ireland for a number of omissions, including: the United Kingdom did not dispute an administrative determination that torture had occurred; a British government committee found the techniques illegal under British law; the United Kingdom had declared it would no longer use the techniques; four judges dissented and wrote separately that they considered the techniques to be torture; and though the majority did not find that the techniques constituted torture, it still found them to be in violation of the European Convention.136 Margolis, the Justice Department reviewer, then rejected OPR analysis on the ground that the agency was doing “little more than . . . substitut[ing] . . . its own judgment for the judgment of Yoo and Bybee on those points.”137

Whatever one thinks of the Torture Memo’s treatment of Ireland, it is striking that Bradbury’s Article 16 Memo never even cited the Ireland decision. Even more striking is the contrast between the Torture Memo’s and the Article 16 Memo’s conclusions. The Torture Memo implied that the CIA’s methods constituted cruel, inhuman and degrading treatment. The Article 16 Memo squarely concluded they did not constitute such treatment—without citing the Torture Memo (which had been withdrawn at that point) and without even a coherent definition of cruel, inhuman and degrading treatment (as was shown above).

OPR would condemn the Torture Memo, but not the Article 16 Memo, as unethical—though the Article 16 Memo was technically and substantively more problematic. OPR investigated Bradbury’s memos, but its Final Report ignored the failure to mention Ireland, the opinion’s arbitrary balancing test, and its disregard of Yoo’s suggestion that the CIA’s methods qualified as cruel, inhuman or degrading.

OPR did identify other problems with Bradbury’s work. In yet another memo, Bradbury had found CIA methods not to be “cruel, humiliating or degrading” under Common Article 3, closely tracking the Article 16

134 Id. at 173. See also id. at 197–98 (discussing the Ireland decision). 135 Id. at 196. 136 OPR Final Report, supra note 8, at 191−92. 137 Margolis Opinion, supra note 6, at 40.


Memo’s analysis.138 OPR wrote that those findings “appear to be inconsistent with the plain meaning and commonly-held understandings of the language of Common Article 3.”139 OPR also found the Article 16 Memo’s discussion “incomplete”140 and “counterintuitive.”141 Moreover OPR noted “several indicia” that Bradbury had written the memos “with the goal of allowing the ongoing CIA program to continue.”142 These “indicia” included reports from other Justice Department officials that Bradbury, who was a candidate for the directorship of OLC when he produced the Article 16 Memo, was being pressured to uphold the interrogation methods.143 OPR also identified “strained” analogies to an American program that aimed to condition soldiers against harsh treatment if they were captured144 and “uncritical acceptance” of CIA representations regarding both the nature of the enhanced interrogation techniques145 and their effectiveness.146

Yet OPR found that these “shortcomings”—including the use of “counterintuitive” legal analysis—did not rise “to the level of professional misconduct.”147 In OPR’s opinion, what separated Bradbury’s work from Bybee’s and Yoo’s was precisely the lawyerly attributes that—as noted in the Introduction—Jack Goldsmith considered critically important. Bradbury’s Memos were “more carefully and thoroughly written” than Yoo’s,148 OPR observed, and they “eschew[ed] reliance on the Commander-in-Chief, necessity, and self-defense sections” that had appeared in the Torture Memo.149 In short, “the memoranda were written in a careful, thorough, lawyerly manner, which we concluded fell within the professional standards that apply to Department attorneys.”150 OPR did

138 OPR Final Report, supra note 8, at 249 (citing the 2007 Bradbury Memo). 139 Id. at 249–50. OPR was referring to a still classified memorandum and its interpretation of

Common Article 3 of the Geneva Conventions, but the same criticism properly applies to the Article 16 Memo.

140 Id. at 250 (referring to both Article 16 and Bradbury Memos’ analysis of substantive due process).

141 Id. at 249. 142 Id. at 241. 143 Id. Bradbury denied, however, that he was pressured in any way, except to complete the

memoranda on time. Id. at 145. 144 OPR Final Report, supra note 8, at 242. 145 Id. at 243. 146 Id. 147 Id. at 251. 148 Id. at 241. 149 Id. at 258. OPR also pointed out that Bradbury “remov[ed] the earlier memoranda’s reliance

on the health benefits statute” and “correct[ed]” the analysis of specific intent. Id. 150 OPR Final Report, supra note 8, at 258.


not explain how an analysis could be “careful, thorough [and] lawyerly” while also being legally counterintuitive.151 But its analysis, as shown below, could have come almost word for word from Goldsmith’s The Terror Presidency.152

3. Persons and Techniques of Interrogation

With the exception of the Torture and Legal Standards Memos, which analyzed the law without reference to specific techniques, each OLC opinion described interrogation methods in detail. No opinion found any method to be illegal; each included a variant of argument #6, above,153 combining an interpretation of the law with a factual judgment about the techniques. OLC construed the phrases “severe pain or suffering” and “prolonged mental harm” and then determined the pain and mental harms associated with interrogations fell short of meeting those standards.

Regarding waterboarding, OPR noted that the opinions ignored a series of military and judicial condemnations of simulated drowning as torture.154 However, OPR would not say that the CIA’s waterboarding technique was the same as those earlier methods of “water torture.”155 Reported cases, according to OPR, did not include “sufficient descriptions . . . to determine how similar the techniques were to those proposed by the CIA.”156 Nor did the earlier cases, “involve[]the interpretation of the specific elements of the torture statute.”157 Although OPR found that Bybee and Yoo failed to provide “thorough, objective, and candid” legal advice about waterboarding, OPR did not deem their advice—or, for that matter, any OLC advice on interrogation—to be wrong.

This difference is important. It is one thing to say that Bradbury

151 Id. See David Cole, The Torture Memos: The Case Against the Lawyers, 56 N.Y. REV.

BOOKS 15 (2009), available at http://www.nybooks.com/articles/archives/2009/oct/08/the-torture-memos-the-case-against-the-lawyers/, for a contrary conclusion about the relative merits of Bradbury’s and Yoo’s memos. Cole harshly criticizes Bradbury, describing the Article 16 Memo as “the most disingenuous of all” and “the May 2005 memos” generally as “the worst of the lot” because they ultimately “reach even more unreasonable positions than the August 2002 memos.” Id.

152 See discussion infra Part IV. 153 Since 2005, interrogation opinions have relied on legal interpretations in the Legal Standards

Memo, which replaced the Torture Memo. Legal Standards Memo, supra note 2 at 1. 154 OPR Final Report, supra note 8 at 234–35 (“The government has historically condemned the

use of various forms of water torture and has punished those who applied it.”). 155 Id. at 235. 156 Id. Cf. Margolis Memo, supra note 6 at 60–61 (finding a description of a waterboard-like

procedure in a previous case, but concluding that Bybee’s and Yoo’s treatment of waterboarding did not amount to professional misconduct). OPR also faulted Bybee and Yoo for not discussing Ashcraft v. Tennessee, 322 U.S. 143, 150, n.6 (1944), which describes “deprivation of sleep” as “the most effective torture.” OPR Final Report, supra note 8, at 236.

157 Id. at 235.


committed legal errors. It is quite another thing to say that he misstated the law and sanctioned torture. Refusing to consider the issue of torture vel non, OPR ensured that Justice Department would remain free of the ethical taint associated with torture.

OPR’s focus on the breadth of legal opinions—and disregard for the reality of interrogations—echoed Jack Goldsmith’s approach. As shown below, Goldsmith’s analyses consistently focus on the broadness or narrowness of opinions and say almost nothing about techniques of interrogation.158 On this basis, both Goldsmith and OPR draw a sharp distinction between pre- and post-Goldsmith OLC opinions. If actual results mattered, however, it would be difficult to distinguish among various OLC opinions, since each upheld every interrogation technique;159 Goldsmith’s and OPR’s legal realism evidently did not extend to questions of breadth. Thus, the first stage OPR analysis found Yoo and Bybee professionally culpable while exonerating Bradbury. OPR’s Report underscored this point, noting that investigators

did not attempt to determine and did not base our findings on whether the [Torture and Military Interrogation memos] arrived at a correct result. Thus, the fact that other OLC attorneys subsequently concluded that the CIA’s use of [enhanced interrogation techniques] was lawful was not relevant to our analysis. Rather, we limited our review to whether the legal analysis and advice set forth in the Bybee and Yoo Memos were consistent with applicable professional standards.160

The result was very questionable judgments by OPR about the

technical and ethical merits of OLC’s opinions.161 Along with its focus on techniques and styles of legal argument, OPR

emphasized very fine factual and conceptual differences: the difference between two methods of using water to induce the sensation of

158 See discussion infra Part IV. 159 See generally Torture Memo, supra note 1; Legal Standards Memo, supra note 2. These

memos, however, did not discuss particular techniques. Goldsmith notes that Levin’s Legal Standards Memo expressed a belief that none of the “conclusions” from “the Office’s prior opinions addressing issues of detainees . . . would be different under the standards set forth in this memorandum.” GOLDSMITH, supra note 20, at 164.

160 OPR Final Report, supra note 8, at 160. 161 Compare GOLDSMITH, supra note 20, at 150–52 (praising the Qaeda Operative Memo as

narrowly written), with OPR Final Report, supra note 8, at 234–37 (finding Bybee and Yoo guilty of professional misconduct because of the memo’s flaws).


drowning.162 OPR resisted generalizations and meaningful abstract categories; rather, every detail mattered. This same tendency marked the later OLC opinions. Bradbury’s memos are filled with minute details about the techniques, with the implication being that every factual detail—every inch of distance that a detainee could move while bound to a chair, for example—was potentially decisive.163 In the world of interrogation, it appeared, every moment and detail was highly individualized and therefore required correspondingly individualized and concrete legal assessments. Splintered and fractionalized, such a world is not conducive to generalizations or abstractions. Thus, OLC imposed an ontological burden, as well as a legal and factual one, on the case for torture.

Indeed, even this ontological burden was individualized. Bradbury readily generalized about the effectiveness of interrogation techniques, for example, and just as readily supposed that combining techniques generally enhanced their effectiveness.164 Yet, he found no reason to think that combining techniques similarly enhanced pains or harms.165 In the Techniques Memo, Bradbury wrote that a “substantial question” existed about whether extended sleep deprivation alone imposed severe physical suffering,166 and also that a “most substantial question” existed about whether waterboarding constituted torture.167 However, despite the fact that each technique was a close call by itself, in the Combined Techniques Memo, Bradbury concluded, that waterboarding could be used simultaneously with sleep deprivation,168 provided there was medical monitoring.169 It was a legal version of Zeno’s Paradox: no matter how close to the torture line you began a journey, and no matter how many

162 See supra text accompanying note 156. 163 E.g. Article 16 Memo, supra note 2, at 110 (describing the number of calories supplied to

detainees during dietary manipulation); id. at 111 (describing the number of feet shackled detainees are able to move); id. (describing the number of hours detainees can be deprived of sleep); id. at 112–13 (describing the temperature of water used in water dousing and maximum length of exposure); id. at 113 (describing the number of seconds a waterboarded detainee will be exposed to the technique); Combined Techniques Memo, supra note 2, at 82–85 (providing a detailed description of a “prototypical interrogation”, and describing a typical waterboarding session in detail).

164 Article 16 Memo, supra note 2, at 110 (“Conditioning techniques are not designed to bring about immediate results. Rather, these techniques are useful in view of their ‘cumulative effect . . . used over time and in combination with other interrogation techniques and intelligence exploitation methods.’”).

165 See, e.g., Combined Techniques Memo, supra note 2, at 88 (noting that CIA physicians confirmed “that the techniques, when combined [in the ways proposed]. . . would not operate in a different manner from the way they do individually, so as to cause severe pain”).

166 Techniques Memo, supra note 2, at 65. 167 Id. at 69. 168 Combined Techniques Memo, supra note 2, at 92. 169 Id. at 95.


steps towards that line you took, you somehow never crossed the line. The Combined Techniques Memo had cautioned that its conclusions

would not extend to detainees or combinations of methods “unlike” those it specifically considered.170 For Deputy Attorney General James Comey, even that did not go far enough. According to OPR, Comey considered the entire memo too “theoretical”; since it was “not tied to a request for the use of specific techniques on a specific detainee . . . it was irresponsible to give legal advice about the combined effects of techniques in the abstract.”171 In effect, Comey postulated a world so ontologically fractured that legal judgments had to be made detainee-by-detainee as well as technique-by-technique. Comey’s view reflected a more radical anti-generalization stance than even Bradbury’s, and it would have limited the amount of torture used in interrogations. Writing OLC opinions for each detainee would consume time, and the limited time available for doing so would restrict the number of interrogations.

One can infer a similarly fractured view of human personality in the OLC opinions—a view radically different from both the original views that underlay CIA interrogation methods and from ordinary lay understanding. Tracing the history of CIA techniques, Alfred W. McCoy described how the agency “fused two new methods, ‘sensory disorientation’ and ‘self-inflicted pain’” to produce a new form of torture.172 According to McCoy,

[t]hrough relentless probing into the essential nature of the human organism to identify its physiological and psychological vulnerabilities, the CIA’s “sensory deprivation” has evolved into a total assault on all senses and sensibilities—auditory, visual, tactile, temporal, temperature, survival, sexual, and cultural. Refined through years of practice, the method relies on simple, even banal procedures—isolation, standing, heat and cold, light and dark, noise and silence—for a systematic attack on all human senses. The fusion of these two techniques, sensory disorientation and self-inflicted pain, creates a synergy of physical and psychological trauma whose sum

170 Id. 171 OPR Final Report, supra note 8, at 141 (characterizing Comey’s position, but not directly

quoting him). The British investigation of the interrogation techniques at issue in United Kingdom v. Ireland was notorious for overlooking the cumulative effect of the various interrogation methods, a development that Comey may well have known about. See MCCOY, supra note 3, at 56 (discussing the “Compton Report”).

172 MCCOY, supra note 3, at 8.


is a hammer-blow to the fundamentals of personal identity.173

Because of the historic framework, it might be expected that McCoy,

an opponent of torture, would invoke “the fundamentals of personal identity.”174 Yet those who developed the techniques described them in similar terms. For example, an early report of a sensory deprivation experiment concluded that “‘variation in the sensory environment . . . would contribute to the breakdown of the organized activity of complex central processes’ in the brain”175—an account perhaps foretold by Beccaria’s reference to “muscles and fibres.”176 Recasting such claims into the language of personality, the CIA’s 1963 handbook for interrogators stated as its “fundamental hypothesis” that interrogation techniques—including methods of sensory deprivation—were, “in essence methods of inducing regression of the personality to whatever earlier and weaker level [was] required.”177

OLC’s opinions recognize no effects of this kind. Rather than producing a “regression” in personality or a “breakdown” in brain processes, interrogation techniques merely “wear down the detainee, physically and psychologically.”178 Water dousing, for example, “is intended to weaken the detainee’s resistance and persuade him to cooperate.”179 In effect, OLC saw objects that could be worn down, but not persons who could be broken. Metaphorically at least, OLC failed to notice the damage interrogations produced because OLC missed the fact that persons were involved. Neither here nor elsewhere, however, did OLC explain how such effects could get produced, psychologically or biologically.

173 Id. 174 Id. 175 Id. at 41 (quoting Woodburn Heron & D.O. Hebb, Cognitive and Physiological Effects of

Perceptual Isolation, in SENSORY DEPRIVATION: A SYMPOSIUM HELD AT HARVARD MEDICAL SCHOOL (Philip Solomon ed., 1961)).

176 See supra text accompanying notes 57 and 61. 177 MCCOY, supra note 3, at 51 (quoting CENT. INTELLIGENCE AGENCY, KUBARK

COUNTERINTELLIGENCE INTERROGATION 41 (1963), available at http://www.gwu.edu/~nsarchiv/ NSAEBB/NSAEBB27/01-01.htm).

178 Combined Techniques Memo, supra note 2, at 88. See also Article 16 Memo, supra note 2, at 112 (noting the CIA considers a particular technique, walling, effective because it “wears down” the detainee and “creates a sense of dread”).

179 Article 16 Memo, supra note 2, at 113 (quoting the Techniques Memo, supra note 2, at 37). See also Techniques Memo, supra note 2, at 39 (noting that the purpose of sleep deprivation is “to weaken the subject and wear down his resistance”).


OLC’s view also departs from ordinary, lay understandings. When a catastrophe, like a murder, occurs at a school or on a college campus, counselors are dispatched to minimize the psychological harm done to bystanders. Imagine, however, if some students were physically taken, hooded, disoriented, confined in a small box, “walled,” deprived of sleep and food and then waterboarded. Further, imagine if counselors were deemed unnecessary for those very same students because long term psychological harm was said to be unlikely (and because physicians, who had watched everything, said that they noticed no signs of harm). Such claims would seem absurd—yet OLC made precisely those claims throughout its “careful” and “thorough” legal analyses.

Using a less fractured ontology, and a more conventional view of persons, critics see obvious torture. David Cole wrote:

OLC lawyers contorted the law to authorize precisely what it was designed to forbid. They concluded that keeping suspects awake for eleven days straight, stripping them naked, exposing them to cold temperatures, dousing them with water, slamming them into walls, forcing them into cramped boxes and stress positions for hours at a time, and waterboarding them hundreds of times were not torture, not cruel, not inhuman, not even degrading, and therefore perfectly legal.180

In this way, OLC’s opinions conform to the historic pro-torture

pattern. They avoid all rights-promoting abstractions, including any concept of personhood. Instead, the opinions rely on minute assessments, unique situations, and weighing or balancing of advantages. Everything is concrete, material and particularized. In anti-torture arguments, individuality inheres in persons; in OLC’s analyses, it resides in facts and circumstances.

Of course, OLC never commented on the nature of persons or offered an explicit policy towards abstractions. Even if I have characterized the opinions accurately, they do not say why they avoided abstractions or what larger purpose, if any, they were attempting to serve. My analysis runs the risk of mistaking normal attributes of a legal opinion for the pattern of a pro-torture argument.

These difficulties disappear, however, with Goldsmith’s memoir of his OLC tenure. Goldsmith explains things that OLC and OPR did not. His memoir strips away the mask of law, and it bares the historic patterns.

180 Cole, supra note 151, at 14.


IV. JACK GOLDSMITH

To recapitulate, two features mark both the post-Goldsmith OLC interrogation opinions and OPR’s Report. First, both place a premium on narrow legal analyses and arguments. Later OLC opinions depart from the Torture Memo largely by declining to state any general position on issues such as presidential power or the available defenses to a charge of torture. Similarly, OPR faulted Yoo and exonerated Bradbury because Yoo offered general legal reasons and Bradbury did not.181 Second, the later interrogation opinions and the OPR Report both manifest an exaggerated orientation to reality. OLC’s opinions are hyper-realistic, emphasizing minute details and studiously avoiding generalizations. OPR’s ethics opinions are the opposite—hypo-realistic—since they consider the accuracy of OLC’s legal conclusions and the reality of interrogation as beside the point. However, the ultimate effect is similar. The interrogation opinions get lost in a maze of detail about duration, distances, and postures while the ethics report flounders in a mass of detailed legal analyses.

These approaches found an early and clear champion in Goldsmith. They were already evident in June 2004, when Goldsmith (along with James Comey and Patrick Philbin) briefed reporters after the Torture Memo was leaked.182 According to the New York Times account of the briefing, the officials “essentially disavowed” the Torture Memo as “overbroad and irrelevant”183 and said it would be “reviewed and revised because it created a false impression that torture could be legally defensible.”184 Already the focus was on “overbroadness” and on the “impression” that the torture memo made, not on actual interrogations or legality.

What has emerged is an alternate pattern of debate, one that only seems to conform to the historic pattern. On one view, the interrogation opinions were deeply flawed because they trampled on fundamental principles and sanctioned torture. On the other view, some interrogation opinions were deeply flawed because they were legally overbroad and gave the impression of sanctioning torture. Confusing these two views, many have hailed Goldsmith as if he were a defender of the traditional approach, rather than the originator of a view that would replace it.

181 See supra text accompanying note 162–63. 182 Published accounts of the briefing did not name Goldsmith, Comey or Philbin, but the OPR

Final Report identifies them. OPR Final Report, supra note 8, at 123. 183 Richard W. Stevenson, White House Says Prisoner Policy Set Humane Tone, N.Y. TIMES,

June 23, 2004, at A1. 184 David Johnston & James Risen, Aides Say Memo Backed Coercion Already in Use, N.Y.

TIMES, June 27, 2004, at N1.


A. Goldsmith at the OLC

Goldsmith’s brief tenure as OLC director was punctuated by serious disagreements with more senior officials. An especially intense dispute concerned the Terrorist Surveillance Program and its warrantless eavesdropping of domestic communications.185 Goldsmith concluded that the program was illegal and persuaded Attorney General Ashcroft to agree.186 Later, with Ashcroft ill, Goldsmith was present in Ashcroft’s hospital room, with Deputy Attorney General Comey and others, during a confrontation in which senior administration officials tried to convince Ashcroft to reverse himself and re-authorize the program from his hospital bed.187 Goldsmith also determined that the Fourth Geneva Convention applied to all Iraqi nationals, even those belonging to Al Qaeda or other terrorist organizations.188

Goldsmith’s principal antagonist was David Addington, Vice President Cheney’s General Counsel and later, the Vice President’s Chief of Staff. Generally agreeing with Addington, although less doctrinaire, was Alberto Gonzales, White House Counsel and later, Attorney General. Goldsmith’s decision about the Geneva conventions, for example, had “puzzled” Gonzales and made Addington “just plain mad.”189 When Goldsmith withdrew Yoo’s March, 2003 Military Interrogation Memo as legally flawed, Addington was “beside himself.”190 Goldsmith’s eventual resignation was associated with his withdrawing the Torture Memo.191

Goldsmith “was astonished, and immensely worried” when he first read the Torture Memo.192 Goldsmith found it “deeply flawed: sloppily reasoned, overbroad, and incautious in asserting extraordinary constitutional authorities on behalf of the President.”193 The Military

185 GOLDSMITH, supra note 20, at 174. 186 Id. 187 MAYER, supra note 44, at 289–90. During Senate testimony, Goldsmith confirmed the

incident. Preserving the Rule of Law in the Fight Against Terrorism: Hearing Before the S. Comm. on the Judiciary, 110th Cong. 14-15 (2007) [hereinafter Rule of Law Hearing].

188 GOLDSMITH, supra note 20, at 40. 189 Id. at 41. See also MAYER, supra note 44, at 264–66 (discussing Goldsmith’s decision and

the reaction of Addington and Gonzales). 190 Daniel Klaidman, Stuart Taylor Jr. & Evan Thomas, Palace Revolt, NEWSWEEK, Feb. 6,

2006, at 36, available at http://www.newsweek.com/2006/02/05/palace-revolt.html. 191 GOLDSMITH, supra note 20, at 161–62. 192 GOLDSMITH, supra note 20, at 10. 193 Id. at 10. This quotation describes Goldsmith’s general reaction to OLC counterterrorism

opinions. Goldsmith’s later discussion makes it clear that he believed the Torture Memo fit that description. See id. at 151 (describing OLC’s analysis as “legally flawed, tendentious in substance and tone, and overbroad”).


Interrogation Memo, in Goldsmith’s view, had the same problems.194 The first flaw in the interrogation memos, according to Goldsmith, lay

in an error of statutory construction.195 In particular, Goldsmith objected to Yoo’s conclusion that the statutory requirement of “severe pain” meant pain commensurate with “organ failure . . . or even death.”196 Yoo had adopted that language from a statute guaranteeing emergency medical treatment at hospitals participating in the Medicare program when treatment would avert, among other things, organ failure or death.197 This was completely unrelated to the definition of torture, Goldsmith noted in The Terror Presidency. Indeed, the medical statute never even defined “severe pain.”198 Yoo’s “clumsy definitional arbitrage,” according to Goldsmith, “didn’t seem even in the ballpark.”199 Goldsmith not only singled out the organ failure analysis as OLC’s “[m]ost notorious” error but cited it to demonstrate that Yoo had “interpreted the term ‘torture’ too narrowly.”200

Yet Goldsmith would not have withdrawn an interrogation opinion because of such “questionable statutory interpretations” alone.201 Other flaws, and particularly Yoo’s treatment of presidential power, were more worrisome. Goldsmith noted that “[m]any prior OLC opinions” had concluded “the President could ignore statutes that in concrete instances conflicted with his commander-in-chief powers.”202 But these earlier

194 Id. at 143. 195 Id. at 144–45. 196 Id. at 145. 197 Torture Memo, supra note 1, at 176 (citing 8 U.S.C. § 1369 (2000); 42 U.S.C. §§ 1395dd,

1395w-22, 1396b, 1396u-2 (2000)). The memo adds that “[a]lthough these statutes address a substantially different subject from Section 2340, they are nonetheless helpful for understanding what constitutes severe physical pain.” Id.

198 GOLDSMITH, supra note 20, at 145. 199 Id. Waldron described perhaps the most trenchant criticism of this aspect of the Torture

Memo. He wrote:

Using these conditions organ failure or death] to define “severe pain” would be like taking the following statement: “A dog (particularly a large dog) is a Dalmatian if it has a white coat with black spots” to imply that the definition of “large dog” required a white coat and black spots.

Waldron, supra note 10, at 1708 n.122. 200 GOLDSMITH, supra note 20, at 144–45 (interestingly, Goldsmith never provided his own

definition of torture). Yoo himself came to regret the organ failure analogy, saying that he “failed to anticipate that the memo would leak and become susceptible to quotations out of context” and that the analysis “did not do justice to the more complete definition [of severe pain] in the memo itself.” YOO, supra note 20, at 177–78. In fact, since severe pain does not accompany every organ failure or death, the analogy does not even work.

201 GOLDSMITH, supra note 20, at 145. 202 Id. at 148.

http://web2.westlaw.com/find/default.wl?tf=-1&rs=WLW10.03&fn=_top&sv=Full&docname=8USCAS1369&tc=-1&pbc=76E4BB2C&ordoc=0305997214&findtype=L&db=1000546&vr=2.0&rp=%2ffind%2fdefault.wl&mt=208

http://web2.westlaw.com/find/default.wl?tf=-1&rs=WLW10.03&fn=_top&sv=Full&docname=42USCAS1395W-22&tc=-1&pbc=76E4BB2C&ordoc=0305997214&findtype=L&db=1000546&vr=2.0&rp=%2ffind%2fdefault.wl&mt=208

http://web2.westlaw.com/find/default.wl?tf=-1&rs=WLW10.03&fn=_top&sv=Full&docname=42USCAS1395W-22&tc=-1&pbc=76E4BB2C&ordoc=0305997214&findtype=L&db=1000546&vr=2.0&rp=%2ffind%2fdefault.wl&mt=208


opinions had addressed “concrete instances”; Yoo, by contrast, concluded that all statutes attempting to “regulate the interrogation of battlefield combatants” violated the Constitution.203 “This extreme conclusion,” Goldsmith wrote,

has no foundation in prior OLC opinions, or in judicial decisions, or in any other source of law. And the conclusion’s significance sweeps far beyond the interrogation opinion or the torture statute. It implies that many other federal laws that limit interrogation . . . [including] the Uniform Code of Military Justice . . . are also unconstitutional, a conclusion that would have surprised the many prior presidents who signed or ratified those laws, or complied with them during wartime.204

In place of this broad approach, the “OLC might have limited its set-aside of the torture statute to the rare situations in which the President believed that exceeding the law was necessary in an emergency, leaving the torture law intact in the vast majority of instances.”205

Goldsmith did not, however, offer his own views about the reach of presidential power. In particular, we do not know how Goldsmith defines “emergency” or what he considers the salient features of those “rare situations.” Nor is clear how often Goldsmith’s judgment about particular cases would differ from Yoo’s. Yoo recognized a general presidential prerogative to control the interrogation of battlefield combatants;206 Goldsmith’s view would lead to different results when such an interrogation did not qualify as emergent. But given the nature of war—and Bradbury’s definition of the relevant government interests—almost all interrogations qualify as emergent.

Another flaw was more general. Yoo’s opinions exhibited a “lack of care and sobriety,” Goldsmith thought, a “tendentious tone . . . lack[ing] the tenor of detachment and caution that usually characterizes OLC work, and that is so central to the legitimacy of OLC.”207 To illustrate, Goldsmith cited the discussion of “defenses and other ways to avoid prosecution.”208 Goldsmith believed that these analyses, “could be

203 Torture Memo, supra note 1, at 207. 204 GOLDSMITH, supra note 20, at 149. 205 Id. at 148. 206 Id. at 187. 207 Id. at 148–49. 208 Id at 149.


interpreted as if they were designed to confer immunity for bad acts.”209 According to Goldsmith, the “final nail in the interrogation opinions’

coffin” was that Yoo “analyzed the torture statute in the abstract, untied to any concrete practices” and reached conclusions “wildly broader than was necessary to support what was actually being done.”210 Indeed, this flaw seemingly encompassed the other two. Yoo’s overly broad treatment of presidential power seemed to be just an example of his characteristically broad approach to legal issues. Similarly, Yoo’s “lack of care and sobriety” produced general rules not limited to particular concrete instances.211

Goldsmith believed that these flaws warranted withdrawing the two memos, despite the “superstrong stare decisis presumption” that generally surrounded OLC opinions.212 However, he also decided not to withdraw them until he could inform the relevant agencies “precisely what interrogation practices were legally available under a proper analysis.”213 In this way, he hoped to “minimize the expected panic” that withdrawing an opinion would cause.214

Carrying out this policy with the Military Interrogation Memo presented no problem. The Defense Department had adopted only “uncontroversial” interrogation methods, none of which came close to constituting torture.215 Accordingly, in December 2003, Goldsmith informed the Pentagon that it could continue to use the same methods of interrogation, but that the Military Interrogation Memo contained “errors” and “should not be relied upon for any reason.”216

The CIA’s methods, waterboarding, stress positions, sensory deprivation, walling and the rest were different. Goldsmith writes:

[I]n contrast to my sense of the Defense Department [interrogation] techniques, I wasn’t as confident that the CIA techniques could be approved under a proper legal analysis. I didn’t affirmatively believe that they were

209 Id. at 150. 210 GOLDSMITH, supra note 20, at 150. 211 Id. at 148. 212 GOLDSMITH, supra note 20, at 146. Goldsmith added that he “knew of no precedent for

overturning OLC opinions within a single administration . . . certainly not on an important national security matter.” Id.

213 Id. at 152. 214 Id. 215 Id. at 154. 216 Id.


illegal either, or else I would have stopped them. I just didn’t yet know.217

Since the Torture Memo undergirded those methods, Goldsmith refrained from withdrawing it and set about working on a replacement.

Goldsmith never completed the replacement memo. He had neither “the time [n]or the resources” to do so because, “[o]ther matters that remain classified, but that everyone in the government agreed were a higher priority, preoccupied [his] time, day and night and weekends, during the first four and a half months of 2004.”218

When the Torture Memo leaked, however, Goldsmith changed course. He decided to withdraw the opinion despite the absence of any replacement and despite his never having determined whether the CIA’s techniques were legal. He explains,

I withdrew . . . [the Torture Memo] even though I had not yet been able to prepare a replacement. I simply could not defend the opinion. I had rejected [the same] . . . reasoning in the March 2003 opinion, and I knew that the [Torture Memo] . . . would eventually suffer the same fate. Delaying the inevitable was only making matters worse, especially since . . . every day the OLC failed to rectify its egregious and now-public error was a day that its institutional reputation, and the reputation of the entire Justice Department, would sink lower yet.219

Goldsmith also submitted his resignation after he “ensure[d] that . . .

withdrawal of the interrogation opinion would stick.”220 Though Goldsmith never produced an interrogation opinion, we are

not in the dark about his views. As noted in the Introduction, Goldsmith praised the Qaeda Operative Memo, even though it had approved waterboarding and other CIA interrogation methods in a particular case.221 Moreover, Goldsmith endorsed the Legal Standards Memo, Daniel Levin’s 2004 replacement for the Torture Memo that reaffirmed all of the CIA’s interrogation techniques. Goldsmith wrote:

217 Id. at 155–56. 218 GOLDSMITH, supra note 20, at 156. 219 Id. at 158. 220 Id. at 161. 221 Id. at 122. See supra text accompanying note 47.


The Levin opinion gave the torture law a much more rigorous and balanced interpretation, correcting the errors and exaggerations of the [Torture Memo]. The new opinion declined to address the presidential override issue analyzed in the earlier memo, reasoning that consideration of these matters “would be inconsistent with the President’s unequivocal directive that United States personnel not engage in torture.” And then, in an important footnote, the Levin opinion stated that “[w]hile we have identified various disagreements with the [Torture Memo], we have reviewed this Office’s prior opinions addressing issues involving treatment of detainees and do not believe that any of their conclusions would be different under the standards set forth in this memorandum.” In other words, no approved interrogation technique would be affected by this more careful and nuanced analysis. The opinion that had done such enormous harm was completely unnecessary to the tasks at hand.222

Thus, Goldsmith’s problem lay not with OLC’s upholding CIA

interrogation methods—waterboarding, stress positions, deprivation of sleep and food and the like—but with upholding them in a way that negatively impacted the Justice Department’s reputation. Goldsmith thought the opinions had done “enormous harm”—but to the OLC, not to the detainees.

B. Goldsmith’s Realism

In The Terror Presidency, Goldsmith described himself as “an improbable choice to lead the Bush Administration’s OLC.”223 He boasted a colorful family background224 and, although politically conservative, had no previous involvement in Republican politics.225 Yet, it is easy to see what made him an attractive candidate. As a law professor at the

222 GOLDSMITH, supra note 20, at 164–65 (internal citations omitted). See also id. at 151

(describing the Torture Memo and the Military Interrogation Memo as “overbroad and thus largely unnecessary”). Mayer suggests that Levin was pressured into reaffirming the interrogation methods, and she finds that reaffirmation “oddly contradictory” with the rest of the Legal Standards Memo. MAYER, supra note 44, at 306–07. Goldsmith’s endorsement suggests, however, that in fact there was no contradiction.

223 GOLDSMITH, supra note 20, at 19. 224 One of Goldsmith’s stepfathers, for example, was “a mob-connected Teamsters executive . . .

who was . . . for decades a leading suspect in [Jimmy] Hoffa’s disappearance.” Id. at 20. 225 Id. at 26.


University of Chicago, Goldsmith became “a leading . . . critic of many aspects of the international human rights movement.”226 Along with John Yoo, he belonged to “a group of conservative intellectuals . . . skeptical about the creeping influence of international law on American law.”227 Later, while working in the Pentagon General Counsel’s office, Goldsmith had pressed those same views.228 OLC directors traditionally boasted excellent academic credentials, and Goldsmith clearly fit the mold. Indeed, he may have seemed likely to follow in the footsteps of his friend John Yoo, who had come to the administration from the Berkeley law faculty. In any event, after Ashcraft blocked Yoo’s promotion, Yoo (or possibly Haynes) recommended Goldsmith for the position.229

Goldsmith came to the agency as a jurisprudential realist about the obligations of the United States under international law, a stance that appealed to Addington and Yoo. Yet Goldsmith did not accept tenets that often define “conservative” theory in constitutional law.230 Addington, Gonzales and Yoo almost surely agreed with Goldsmith’s realism in international law, but they considered terrorism issues to be questions of presidential power and, therefore, matters of constitutional law. In all likelihood, they did not imagine Goldsmith carrying his bona-fide conservative realism from the international law to the constitutional realm. However, that is what he did.

1. The Indeterminacy of Legal Rules

Goldsmith’s approach to legal indeterminacy reflected widely held academic views. Unlike strict realists of the 1930s, he recognized that some laws possess a definite meaning, independent of context or culture. When that definite meaning existed, it provided the answer to legal questions. In many other cases, however, the meaning of law was indefinite. Alluding to both situations, Goldsmith wrote that his job “was to make sure the President could act right up to the chalk line of legality. But even blurry chalk lines delineate areas that are clearly out of bounds.”231

226 Id. at 59. 227 Id. at 21. 228 Id. at 59. 229 OPR Final Report, supra note 8, at 110 n.83 (noting that Goldsmith confirmed this account). 230 A conservative theory of constitutional law is often defined as a belief in determinate legal

rules and an insistence on sharply distinguishing between law and politics. Goldsmith’s non-acceptance of these tenets may be related to his background in international law—although Yoo, it should be noted, shared the same background but had a very different constitutional philosophy.

231 GOLDSMITH, supra note 20, at 78.


2. Law and Politics: The President’s Agenda

When confronted with legal indeterminacy, a conventional legal strategy is to guess how a court would decide the question, even while understanding that the court’s answer is not strictly determined by the law. Obviously, that strategy would not work in a court of last resort; it would require the court to decide based on the court’s prediction of how it would, itself, decide the case. Nor was this the strategy prescribed by Goldsmith for lawyers at OLC. Those lawyers, according to Goldsmith, should not be “neutral to the President’s agenda.”232

Goldsmith noted that, like his predecessor, he was appointed OLC director because he appeared to “share[] the basic assumptions, outlook, and goals of top administration officials.”233 “Philosophical attunement with the administration is legitimate,” Goldsmith added, “because OLC ‘serves both the institution of the presidency and a particular incumbent.’”234 It followed that an OLC lawyer should never push against the administration agenda. And Goldsmith deplored government attorneys who “use legal review as an opportunity to push their [own] beliefs about the appropriateness of [a] proposed action, or to serve the institutional interests of their bureaucracy.”235 Instead, OLC lawyers should push the President’s agenda, at least within the “chalk lines” of the law.236 Goldsmith recalled Elliot Richardson’s observation that, “[a]dvice to a President needs to have the political dimension clearly in view, without a regard for any pejorative attached to the world political”237—and Goldsmith added that “[h]aving the political dimension in view means that OLC is not entirely neutral to the President’s agenda” and that he (Goldsmith) “would work hard to find a way for the President to achieve his ends,” particularly in connection with national security.238

Thus, Goldsmith did not interpret the law using his best professional judgment, knowing that his own philosophical commitments were operating beneath the surface. Rather, he shaped his legal judgments with

232 Id. at 34–35. 233 GOLDSMITH, supra note 20, at 35. At the same time, however, Goldsmith notes that “the

office has developed powerful cultural norms about the importance of providing the President with detached, apolitical advice, as if OLC were an independent court inside the executive branch.” Id. at 33. Goldsmith explains the need to balance these competing considerations of the office, “to preserve its fidelity to law while at the same time finding a way, if possible, to approve presidential actions.” Id. at 38–39.

234 Id. at 34 (quoting unnamed Clinton OLC veterans). 235 Id. at 93. 236 GOLDSMITH, supra note 20, at 78. 237 Id. at 34 (internal quotations omitted). 238 Id. at 35.


administration priorities in view. Politics, “without regard for any pejorative,”239 filled the large interstices created by legal indeterminacy.

Goldsmith’s description of this posture as “legitimate” perhaps makes it more plausible that the administration considered his political philosophy when appointing him. Yet that hardly entitled Goldsmith to consider his own “philosophical attunements” when he interpreted laws. Those “attunements” should have been invisible to Goldsmith, since they are his own—and not something to think about while drawing legal conclusions. And if he had tried to bring his “attunements” to consciousness, so that he could take them into account, it should have been for the purpose of counteracting their influence on his thinking, not in order to “attune” the law to his own, or to the President’s, philosophy. Goldsmith argued that OLC directors in other administrations, including President Clinton’s and President Franklin Roosevelt’s, took his approach.240 Goldsmith quoted Robert Jackson who, when serving as Roosevelt’s Attorney General, said that the Attorney General should not “act as a judge and foreclose the Administration from making reasonable contentions.”241 Meanwhile, during the Clinton administration OLC lawyers “tried to moderate what they perceived as their Republican OLC predecessors’ overly aggressive conception of presidential power,” but

all OLC lawyers and Attorneys General over many decades – were driven by the outlook and exigencies of the presidency to assert more robust presidential powers, especially during a war or crisis, than had been officially approved by the Supreme Court or than is generally accepted in the legal academy or by Congress.242

3. Objective Legal Advice: OLC as a Court

During his confirmation hearings, Goldsmith had committed himself to “continue the extraordinary traditions of the [OLC] in providing objective legal advice, independent of any political considerations.”243 Once in

239 See supra text accompanying note 237. 240 GOLDSMITH, supra note 20, at 35–37. 241 GOLDSMITH, supra note 20, at 35 (internal citations omitted) (quoting Robert H. Jackson, in

EUGENE C. GEHART, AMERICA’S ADVOCATE: ROBERT H. JACKSON 221 (1958)). Jackson further noted however that “he was not ‘quite as free to advocate an untenable position because it happens to be his client’s position as he would if he were in private practice,’ because ‘he is the legal officer of the United States’ and has a ‘responsibility to others than the President.’” Id. at 35. See also infra Part IV.B.4 (contrasting Jackson’s opinions with Goldsmiths commentary).

242 GOLDSMITH, supra note 20, at 36–37. 243 Id. at 34.


office, however, he found it “hard” to live up to that commitment.244 (Indeed, as just described, Goldsmith no longer really wanted to.) He had come to believe that the idea of “objective legal advice” was more complicated than he had supposed.

“There is no magic formula,” he writes, “for how to combine legitimate political factors with the demands of the rule of law.”245 Yet The Terror Presidency suggests a branch-specific concept of legal objectivity. On this view, “objective legal advice” within the executive branch need not qualify as strictly “objective” anywhere else, and particularly not in the courts. Goldsmith writes:

Legal advice to the President from the Department of

Justice is neither like advice from a private attorney nor like a politically neutral ruling from a court. It is something inevitably, and uncomfortably, in between.

OLC also needn’t look at legal problems the way courts do. Most Americans (including most lawyers) think the law is what courts say it is, and they implicitly equate legal interpretation with judicial interpretation. But the executive branch does not have the same institutional constraints as courts, especially on national security issues where the President’s superior information and quite different responsibilities foster a unique perspective. In addition, for many issues of presidential power there are no controlling judicial precedents.246

Like the leap from legal indeterminacy to implementing the President’s

agenda, the supposed autonomy of presidential law does not really follow from the considerations that Goldsmith cites. The President’s “superior information” need not entail any different way of looking at problems. Moreover, courts can acquire the same information if the President will produce it in judicial proceedings. Further, while the President surely has “different responsibilities” and a “unique perspective” than courts do, the existence of non-judicial responsibilities and perspectives does not generally entitle litigants to create their own unique, “objective” legal realities.

At another point in his book, Goldsmith explicitly likens OLC to “an

244 Id. 245 Id. at 38. 246 Id. at 35–36 (citation omitted).


independent court inside the executive branch.”247 He expanded on the idea in his interviews with OPR, telling investigators that he had taught a course called “Lawyering for the President” at Harvard and that a “debate” existed over the proper “interpretative stance” for OLC.248 There were “multiple questions,” Goldsmith said:

To what extent should OLC be trying to give neutral,

independent court-like advice, or should OLC be more like giving an attorney’s advice to a client about what you can get away with and what you are allowed to do and what your risks are, [or] something in between. What are the sources of interpretation? Is OLC bound by Supreme Court decisions? . . . [C]an the Executive Branch take an independent role in interpreting the Constitution and the statutes? . . .

[O]ne day I’m going to write a book and they’re difficult questions . . .

But, as a general matter . . . with all those caveats . . . I think the answer is that it is clear that OLC is supposed to serve some independent role within the Executive Branch to try to provide independent advice.249

While the interview largely restates Goldsmith’s position in The Terror

Presidency—often in the same words—there was an interesting difference. In the book, the alternatives for OLC’s opinions were “a politically neutral ruling from a court,” “advice from a private attorney,” or “something . . . in between.”250 In the interview, the alternatives were “neutral, independent court-like advice,” “advice to a client about what you can get away with,” or “something in between.”251 There, Goldsmith’s answer was that OLC had to play “some independent role within the Executive Branch to try to provide independent advice.”252 What changed? The book spoke of a “politically neutral ruling from a court”; the interview, of “neutral, independent court-like advice.” The difference is subtle, but the “neutral advice” in the interview is no longer explicitly neutral politically. Further, in the interview, the alternative of a “ruling from a court” becomes

247 Id. at 33. This idea is not unique to Goldsmith. See supra note 38–39 and accompanying text.

248 OPR Final Report, supra note 8 at 18. 249 Margolis Opinion, supra note 6, at 18. 250 GOLDSMITH, supra note 20, at 35. 251 Margolis Opinion, supra note 6, at 18. 252 Id.


“independent court-like advice.” These changes suggest that Goldsmith no longer thought that OLC’s ignoring the President’s political agenda, and giving advice based purely on a prediction of how courts would interpret law, was even possible.

4. A Faux Framework

Goldsmith’s subtle recasting of OLC’s role is an illustration of the anti-torture chimera that emerges from his work. Reading the OPR interview, it is easy see Goldsmith as a defender of the “rule of law” who would provide “independent” and “neutral” advice about the legality of CIA interrogations. However, after reading the interview carefully, and comparing it with Goldsmith’s comments in his book, it appears that “independent,” “neutral,” and “court-like advice” does not mean advice that attempts to conform with a predictable court ruling. Nor does it denote advice that is independent of the President’s political agenda. Such things are no longer conceivable options. “Independent,” “neutral,” and “court-like” all have new and subtly different meanings, all slanted toward presidential power. The “rule of law” is no longer independent of presidential agendas.

Goldsmith is a talented writer, and one might think that what I have called a “chimera” is no more than the light that any good lawyer can cast on a problem. But something deeper is at work. For another example, consider Goldsmith’s discussion of Robert Jackson’s views. Goldsmith wrote in The Terror Presidency:

What Robert Jackson said fifty-five years ago was still true during my time in office: “a judge, like an executive adviser, may be surprised at the poverty of really useful and unambiguous authority applicable to concrete problems of executive power as they actually present themselves.” When OLC writes its legal opinions supporting broad presidential authority in these contexts as OLCs of both parties have consistently done they cite executive branch precedents (including Attorney General and OLC opinions) as often as court opinions. These executive branch precedents are “law” for the executive branch even though they are never scrutinized or approved by courts.253

Quoting Justice Jackson’s Steel Seizure opinion, Goldsmith seems not to notice that Jackson’s substantive positions were the opposite of his own.

253 GOLDSMITH, supra note 20, at 36 (citation omitted).


Far from giving the President’s “agenda” a preferred legal status, Jackson’s opinion expressed concern about presidential “dictatorship,”254 roundly rejected the idea of “necessity” or “crisis” creating presidential power,255 and dismissed Jackson’s own prior statements as the “self-serving press statements of [an] attorney for one of the interested parties . . . .”256 While Jackson famously observed that the Constitution diffuses political power among the three branches, he in no way suggested that each branch possessed its own power to determine law. To the contrary, Jackson worried intensely about the vagaries of political power becoming vagaries of law.257 Goldsmith, however, seemingly fails to see the differences.

Goldsmith touches on Robert Jackson in another connection. The Terror Presidency praises Jackson’s opinion as Attorney General regarding the “destroyers for bases” deal with Great Britain in 1940.258 At first, President Roosevelt believed that the exchange required Congressional approval. Yet Congress seemed unwilling. Considering the destroyers indispensable to Great Britain’s survival, Roosevelt decided to proceed without Congressional approval.259 Goldsmith lauds the political measures Roosevelt undertook, including a public relations campaign and consultations with political leaders from both parties.260 Roosevelt did not stop there, however. He also obtained an opinion from Jackson upholding the arrangement. Here is The Terror Presidency’s account:

Jackson[’s opinion] got around . . . problems that for months the President and almost everyone else had thought made the transfer impossible . . . [using] legal sleights of hand that were criticized at the time as flawed, but that Jackson concluded were reasonable enough as the bombs rained down on London . . . . Jackson’s opinion was written in a way to minimize the sting of what many viewed as a dangerous assertion of unilateral presidential power. It had a sober tone and was written narrowly so as not to approve one thing more than it needed to. It

254 Youngstown Sheet & Tube Co. v. Sawyer (Steel Seizure), 343 U.S. 579, 653 (1952) (Jackson,

J., concurring). 255 Id. at 646. 256 Id. at 647. 257 See Korematsu v. United States, 323 U.S. 214, 244 (1944) (Jackson, J., dissenting) (“[I]f we

cannot confine military expedients by the Constitution, neither would I distort the Constitution to approve all that the military may deem expedient.”).

258 GOLDSMITH, supra note 20 at 199. 259 Id. 260 Id. at 197–98.


declined to rely on the President’s commander-in-chief powers and focused instead on what Congress had approved and prohibited. It went out of its way to acknowledge that the President’s power over foreign relations “is not unlimited.” And as if to prove the point . . . Jackson concluded that Congress had in fact banned a related transfer of mosquito boats then under construction, and that the President was bound by this determination.261

Goldsmith vests Jackson’s opinion and Roosevelt’s actions in the

destroyer deal with every feature that he found wanting in the Torture Memo. Like Jackson’s legal handiwork, the Torture Memo should have had “a sober tone” and should have been “written narrowly.” Like Jackson himself, Bybee and Yoo should have “decline[ed] . . . to rely on the . . . commander-in-chief’s powers” and should have “acknowledge[d] that the President’s power over foreign relations ‘is not unlimited.’”262 The Bush administration should have followed Roosevelt’s example instead of aggressively claiming broad executive powers.

In the OPR interview, Goldsmith described Jackson’s analysis of the “destroyer for bases deal” as “very bad” and technically “terrible.”263 But Goldsmith’s account of the destroyer deal squares poorly, however, with his metaphor of a legal “line” that OLC lawyers cannot cross. Jackson wrote a “terrible” opinion that respected no legal lines, yet it also constituted outstanding executive lawyering, in Goldsmith view, an exemplar for the War on Terror. With Goldsmith’s legal lines drawn in “chalk,” presidents apparently can erase them and redraw them at will. These redrawn lines, according to Goldsmith, should resemble the old lines as much as possible, a feat that may require “legal sleight[] of hand.”264 What matters is that the President and the executive lawyers appear accommodating and self-effacing, even as the lines move.

The trouble is that this account completely undercuts OLC’s credibility. If we believe it, why should we accept any argument from executive branch lawyers in a national security matter—including the argument that “enhanced interrogation techniques” did not constitute torture? OLC conceivably tried the same “sleight[] of hand” that Jackson

261 Id. 262 Id. at 198. 263 Margolis Opinion, supra note 6, at 19. Goldsmith went on to argue that Bybee and Yoo

should be subject to the same standard as Jackson would be, and not be disciplined. Id. (“And I guess in my preaching moments I would say whatever standard you bring to bear here, it should apply to Justice Jackson . . . as well.”).

264 GOLDSMITH, supra note 20, at 198.


did.265 Perhaps the most interesting part of Goldsmith’s OPR interview

concerned his own role. Comparing enhanced interrogations and John Yoo in 2002 with destroyers and Robert Jackson in 1940, Goldsmith identified a critical difference, namely, himself. Goldsmith said, “I guess . . . the difference . . . is my fault. The difference here from . . . Justice Jackson’s opinion. . . is that you had someone in the office [Goldsmith] say no, those opinions were wrong. So, you’ve got opinions where I say these are in some respects erroneous.”266

This suggests that Goldsmith considered the Torture Memo “terrible” in the way that Jackson’s opinion was terrible. Yet, that was not the case. Although Goldsmith did identify some legal errors, what he objected to most of all was the Torture Memo’s tone and supposed overbreadth. But Jackson’s legally “terrible” opinion was not overbroad. In fact, Goldsmith’s problem with Bybee and Yoo is that they failed to emulate Jackson. Had Goldsmith been a member of Justice Department in 1940, he presumably would have praised Jackson’s effort, not criticized it.

Goldsmith’s accepting “fault” in the interview seems to reflect genuine confusion on his part. He recognizes a realm of conventional law controlled by courts, and a realm of executive law influenced by the President’s agenda. He identified executive-type, essentially political flaws in the Torture Memo—flaws that made it vulnerable to public criticism. Yet, he characterized the Torture Memo’s flaws as legal in nature, as if OLC belonged to the conventional legal realm, where the “rule of law” is not infused with political considerations. Jackson’s opinion was brilliant in the realm of executive law and “terrible” in the realm of conventional law—and Goldsmith two accounts of the opinion seemingly treat the two realms as if they were the same. Thus, Jackson could be a legal model for the Bush administration one day, and the purveyor of “terrible” law on another.

With Goldsmith possibly confused himself, it is small wonder that commentators place Goldsmith in the conventional legal realm too. Yet his “objective legal advice” is not their objective legal advice. Nor is his “rule of law” their rule of law. Perhaps for just this reason—because it can be mistaken for the historic framework—Goldsmith’s approach is shaping contemporary debate about lawyers, interrogation, and torture.

265 Id. 266 Margolis Opinion, supra note 6, at 19.


C. Goldsmith’s Minimalism

A legal “minimalist,” according to Cass Sunstein’s definition, “avoid[s] broad rules and abstract theories,” favoring narrow solutions to specific problems instead.267 The Terror Presidency is a perfect example. It considers a range of legal issues but advances no rules or abstract legal theories of its own. Readers never learn how Goldsmith would define torture, for instance, or when he thinks a president can constitutionally disregard a statute. The book is a model for leaving things unsaid, as, indeed, was Goldsmith’s tenure at OLC, which produced no opinions on the interrogation methods. In matters of interrogation, Goldmisth as OLC director was the living embodiment of legal minimalism.

Circumstances helped. The CIA interrogation techniques remained secret when The Terror Presidency appeared. Thus, Goldsmith could say nothing about them in the book.268 Had waterboarding, walling, and sustained stress positions appeared in The Terror Presidency, it would have been much harder for Goldsmith to emerge as a champion of the rule of law.

Goldsmith’s principal criticism of the Torture Memo was a minimalist one. As noted earlier, Goldsmith considers the “final nail” in the memo’s “coffin” to be that it “analyzed the torture statute in the abstract, untied to any concrete practices” and that it reached conclusions “wildly broader than was necessary to support what was actually being done.”269 Conversely, Goldsmith highlights the Qaeda Operative Memo as “hyper narrow and cautious and splitting hairs and not going one millimeter more than you needed to answer the [legal] question.”270 The same themes run through Goldsmith’s book.

In substance, then, Yoo’s problem was that he was not a jurisprudential minimalist. Goldsmith objected to how much Yoo said, rather than to what Yoo had authorized. The opposite of a minimalist, Sunstein explained, is a “maximalist”: namely, someone who “seek[s] to decide cases in a way that

267 CASS R. SUNSTEIN, ONE CASE AT A TIME: JUDICIAL MINIMALISM ON THE SUPREME COURT 9

(1999). 268 Goldsmith’s book required security clearances. GOLDSMITH, supra note 20, at 12. In Senate

testimony Goldsmith explained that “by contract and law” he was bound to withhold from the Committee even his own legal analysis of the Terrorism Surveillance Program. Rule of Law Hearing, supra note 187, at 10. Goldsmith also testified that pre-clearance had produced no changes in his manuscript other than “two words” being removed at “someone[‘s]” request and aliases being substituted for some names.” Id. at 25.

269GOLDSMITH, supra note 20, at 150. See also supra text accompanying note 210. 270 OPR Final Report, supra note 8, at 122. See also supra text accompanying note 47.


sets broad rules for the future and that also gives ambitious theoretical justifications for outcomes.”271 Sunstein’s maximalist was a perfect description of John Yoo.

The main conceptual impetus for minimalism, at least on Sunstein’s account, is distrust of generalization.272 Minimalism would suit a world in which reasons mattered little and immediate outcomes counted for almost everything. Viewed in the abstract, this makes minimalism a perfect match for those who would find no broad significance in concepts like “rule of law” or “personhood.”

1. Substantive Commitments: The Balance

Sunstein recognized a rich array of general constitutional rights, but he did so by positing the existence of a widely accepted constitutional “core.” Although he called these core rights “minimalism’s substance,” the rights themselves did not derive from minimalism. Instead, Sunstein simply took them as givens. The result was that Sunstein subscribed to an array of rights that looked typical of a modern constitutional liberal’s. Minimalism, for him, meant largely that he would be cautious about extending or expanding those rights.273

Sunstein suggested that different varieties of minimalism exist, each with different substantive commitments.274 Certainly, Goldsmith’s minimalism does not have the same “core” or “substance” as Sunstein’s.275 Apart from the President’s agenda, Goldsmith discusses what might be described as a substantive commitment at only one point in his book. There, Goldsmith observes that it was “unusually important for OLC to

271 SUNSTEIN, supra note 267, at 9–10. 272See Sheldon Gelman, The Hedgehog, the Fox, and the Minimalist 89 GEO L.J. 2297, 2299

(2001) (reviewing SUNSTEIN, supra note 267) (arguing that “[a]n antigeneralization principle lies at the heart of Sunstein’s [concept of]. . . ‘minimalism’”). See also SUNSTEIN, supra note 267, at 10–11.

273 SUNSTEIN, supra note 267, at 61–72. For further discussion see Gelman, supra note 272, at 2314 (“Sunstein uses the phrase `minimalism’s substance’ to describe principles that are not distinctively minimalist.”).

274 Sunstein wrote:

[I]t is easy . . . to imagine minimalists of many different substantive stripes. A conservative minimalist might try to preserve what is essential to conservatism, but leave many important issues undecided . . . . A minimalist who is also a deliberative democrat would attempt to promote the “core” of that commitment.

We could even imagine liberal minimalism, socialist minimalism, Ku Klux Klan minimalism, libertarian minimalism, Aristotelian minimalism, communist minimalism, Nazi minimalism, and so forth.

SUNSTEIN, supra note 267, at 61–62. 275 Sunstein himself, as noted earlier, was highly critical of the Torture Memo. See, e.g., Liptak,

supra note 8.


provide careful and sober legal advice about the meaning of torture” because of the “unusually high” stakes involved—stakes that made flaws in Yoo’s analysis “unusually worrisome.”276 Goldsmith described these “stakes” as follows:

On the national security side of the balance potentially stood tens of thousands of lives, economic prosperity, and perhaps our way of life. On the other side of the balance lay the United States’ decades-long global campaign to end torture, relations with the Muslim world, and the nation’s moral reputation and honor.277

Realistically, these scales could tilt only one way: in the direction of

lives, prosperity and harsh interrogation. No conceivable damage to reputation could outweigh those things. Indeed, since the interrogation methods did not in fact constitute torture—again, Goldsmith endorsed Levin’s views—any damage to reputation would be unwarranted. Such scales do not weigh “apples against apples” or even “apples against oranges”—rather, they weigh apples against mistaken ideas about apples.

Goldsmith’s characterization is revealing in a more subtle way. In historic debates about torture, “way of life”278 would seem to belong on the anti-torture side, aligned with concepts like national identity or the rule of law. Goldsmith, however, places “way of life” on the side of harsh interrogation, grouping it with the physical existence of citizens and with material prosperity. In isolation, “way of life” sounds like an abstraction, but Goldsmith links it with things material and concrete.279 Clearly, Goldsmith is not arguing that waterboarding deepens or enhances our way of life. Rather, “way of life” simply constitutes the sum total of how we live.

Beccaria had said that torturers separate persons from their human nature, seeing only “muscles and fibres.”280 Here, Goldsmith does something similar for nations, associating national “way of life” with the physical existence of citizens and their material well being. But Goldsmith’s trope is more complex than Beccaria’s. Besides transforming “way of life” into something material, it replaces classic anti-torture abstractions, like national identity and the rule of law, which Goldsmith

276 GOLDSMITH, supra note 20, at 146, 148. 277 Id. at 148. 278 See, e.g., DARIUS REJALI, TORTURE AND DEMOCRACY xvi (2007) (invoking an autocratic

“way of life” that features a readiness to torture opponents). 279 Id. 280 See supra text accompanying notes 57 and 61.


does not mention here, with the non-political entities “moral reputation and honor.”281 However, “moral reputation” and “honor” lack the complex conceptual make-up of the anti-torture concepts they replace. In this way, the abstract, argumentative trumps of classic anti-torture argumentslose their quality as trumps and become instead mere air on one side of a balance scale—on the side opposite to “thousands of lives” and our “way of life.”

2. Institutional Commitments: OLC

Sunstein’s minimalism has an overriding institutional objective, which was to preserve the Supreme Court’s legitimacy and role in national life.282 Minimalist decisions, Sunstein believed, would accomplish that, but broad decisions would not. Goldsmith’s minimalism has a parallel purpose: preserving OLC’s role and legitimacy.

Yoo’s cardinal failure, as Goldsmith sees it, was producing opinions that damaged OLC’s reputation when they became public. Goldsmith declaimed that “[t]he opinions lacked the tenor of detachment and caution . . . that is so central to the legitimacy of OLC.”283 Concern over OLC’s reputation, as we have seen, led Goldsmith to withdraw the Torture Memo despite the lack of a replacement for it. “[I]t had become apparent,” Goldsmith writes, “that every day the OLC failed to rectify its egregious and now public error was a day that its institutional reputation, and the reputation of the entire Justice Department, would sink lower yet.”284 Thus, even though national reputation carried little weight on Goldsmith’s balancing scales, the reputation of OLC was decisive.285

Goldsmith never explains, however, why the public cared when OLC appeared to sanction torture. Why did the public care about torture? In reality, public reaction was driven by the abstractions underlying the framework. Ignoring those abstractions, however, Goldsmith perceives only potential reputational problems. Concepts like humanity or the rule of

281 See supra text accompanying note 277. 282 Gelman, supra note 275, at 2346 (arguing that in Sunstein’s view only the Court’s

calculations about its own capabilities and prospects for self-preservation limit its authority). 283 GOLDSMITH, supra note 20, at 149. 284 Id. at 158. Goldsmith was also concerned that the broad legal assertions in Yoo’s opinions in

the Torture Memo and Military Interrogation Memo might be taken as a “green light to justify interrogations much more aggressive than ones specifically approved.” Id. at 151. His solution was that OLC issue an instruction requiring that new interrogation methods receive specific approval from OLC. Thus, Philbin had told the Defense Department that it could not rely on the Military Interrogation Memo to approve new methods. OPR Final Report, supra note 8, at 111. Goldsmith conveyed a similar instruction when he withdrew that memo, telling the Defense Department that the interrogation methods currently in use were “legal” but that the Department should obtain “additional legal guidance [from OLC] before approving any other technique. GOLDSMITH, supra note 20, at 154.

285 See discussion of “argumentative trumps” supra Part II.B.


law become mere flash points that could trigger a public reaction. Where anti-torture critics attack the Torture Memo for undermining the rule of law, Goldsmith attacks it for undermining the OLC. And some commentators, as we have seen, failed to notice the difference.

V. GOLDSMITH AND THE HISTORIC DEBATE

A. Goldsmith and Beccaria

Goldsmith’s views are easy to locate within the historic framework of debate. Concepts and abstractions that traditionally appear in anti-torture arguments—national and personal identity, civilization, humanity, the rule of law—are lacking in The Terror Presidency. Instead, we find the arguments and entities characteristic of the opposite side of the debate: material things, maximizing calculations, and balancing. Goldsmith purports to “weigh” those material things against national reputation.

For Beccaria and Voltaire, humanity and human rights were the fundamental anti-torture abstractions. The Terror Presidency does cite a “human rights culture” at work in the world, but Goldsmith condemns it.286 That culture incubates and nurtures the “human rights industry,”287 he writes, an industry that attacks American interests using “lawfare” which has been defined as “the strategy of using or misusing law as a substitute for traditional military means to achieve an operational objective . . . .”288 Employing lawfare, the human rights industry attempts to hail the United States and American leaders before international tribunals, including the International Criminal Court,289 which Goldsmith describes as “at bottom an attempt by militarily weak nations . . . to restrain militarily powerful nations.”290

Indeed, Goldsmith does not seem to recognize “human rights” as an independent entity. He writes of the human rights “culture,” of the human rights “industry” and of international human rights “law” as undesirable. Of “human rights” themselves, however, Goldsmith says nothing. Further, the concept of “humanity” is simply missing from The Terror Presidency.

Goldsmith’s memoir does describe an incident, however, in which he felt sympathy for the humanity of a detainee. On Goldsmith’s fortieth

286 See generally GOLDSMITH, supra note 20, at 53–64. Goldsmith describes himself as “leading academic critic of many aspects of the international human rights movement.” Id. at 59.

287 Id. at 59. 288 Id. at 58 (quoting Charles J. Dunlap, Jr., Address at the Air and Space Conference and

Technology Exposition: The Law of Armed Conflict (Sept. 13, 2005), available at http://www.afa.org/ Media/scripts/Dunlap_conf2005.asp).

289 Id. at 61–62. 290 Id. at 63.


birthday, he was part of a group touring detention facilities. At Guantanamo Bay, some detainees “looked at us with an anger that I had never before experienced.”291 Returning to Washington, Goldsmith’s group stopped at a South Carolina military prison and saw Jose Padilla, an American citizen being held at that point as an enemy combatant. They then visited the Virginia military prison housing Yaser Hamdi, whose lawsuit would eventually produce a Supreme Court decision against the administration. Goldsmith writes:

Top administration lawyers crowded around the small

black-and-white closed circuit television bolted in the back corner of the room, and witnessed the barely twenty two-year old Yaser Hamdi—it was his birthday as well—in the corner of his small cell in an unused wing of the brig, crouched in a fetal position, apparently asleep.

Before I saw him on the closed-circuit television, I had no sympathy for Hamdi, whom I knew had volunteered to fight for the tyrannical Taliban. Witnessing the unmoving Hamdi on that fuzzy black-and-white screen, however, moved me. Something seemed wrong. It seemed unnecessarily extreme to hold a twenty-two-year-old foot soldier in a remote wing of a run-down prison in a tiny cell, isolated from all human contact and with no access to a lawyer. “This is what habeas corpus is for,” I thought to myself, somewhat embarrassed at the squishy sentiment . . . . I immediately thought my reaction was misplaced. I didn’t question the legality of holding Hamdi. I had no doubt that (as the Supreme Court would affirm twenty months later) the administration had legal authority to detain [him] . . . . My real thought was whether it was prudent to do so in this way, in these circumstances.

That fleeting qualm on my fortieth birthday was my first insight into a characteristic mistake that the Bush administration made in the war on terrorism. On issue after issue, the administration had powerful legal arguments but ultimately made mistakes on important questions of policy. It got policies wrong, ironically, because it was excessively legalistic, because it often substituted legal analysis for political judgment, and because it was too committed to expanding the President’s

291 Id. at 100. Goldsmith’s group also witnessed an interrogation, which he does not describe.


constitutional powers.292 This passage is perhaps the most evocative and realistic in The Terror

Presidency. Elsewhere Goldsmith views lawyers and law through the lens of an abstract legal philosophy, albeit one that purported to focus on the “concrete.” Here, administration lawyers are literally looking at Hamdi through a non-legal lens. Goldsmith’s and Hamdi’s shared birthday bespoke their common humanity. Yet the same passages throw into even sharper relief Goldsmith’s view that humanity should have no bearing on law. Goldsmith artfully depicts a “concrete instance” calculated to convince anyone that humanity mattered, and then dismissed his reaction as a “fleeting qualm on my . . . birthday” and an “embarrass[ing] . . . squishy sentiment.”293 What Goldsmith saw was politically germane, he believes, but legally and morally irrelevant. The sight prompted Goldsmith to think about political “prudence” and nothing more.

Goldsmith is a realist, as we have seen, and he does not believe that legal interpretations rest on law alone. His own interpretations were influenced by the administration’s security agenda. He devised a balancing test that vaunted national security and he described OLC’s responsibilities as “combin[ing] legitimate political factors with the demands of the rule of law.”294 But if legal interpretations properly consider the administration’s political agenda and national security imperatives, why not considerations of humanity as well? Putting humanity out of bounds, Goldsmith had eliminated the heart of classic arguments against torture.

B. Ironies and Contradictions

Goldsmith’s account yields serious ironies and contradictions. The principal difficulty with the Torture Memo, he argues, is the damage it caused to OLC’s reputation. Goldsmith makes the argument, however, in a book that calls on OLC attorneys to “push” law in accord with the administration’s agenda,295 that praises legally “terrible” opinions because they advanced a President’s policy,296 and that employs a balancing test clearly calculated to uphold harsh interrogations.297 In purely political terms, those things seem far more damaging to OLC’s reputation than

292 GOLDSMITH, supra note 20, at 101–02 (footnote omitted). 293 Id. at 102. 294 Id. at 38. 295 See supra text accompanying notes 234–36. 296 See supra pages 162–64. 297 Id. at 38–39.


anything in the Torture Memo. Further Goldsmith misread the public outcry over the Torture Memo.

Yoo’s “organ failure” analysis, depiction of criminal defenses, and claims of expansive Presidential power all aroused sharp public criticism, just as Goldsmith says. But Goldsmith ignores an essential component of the public reaction, namely, revulsion at the actual interrogation techniques OLC had approved.298 The public, unlike Goldsmith, was concerned with what interrogators did.

Again, public concerns were expressed about OLC’s independence from the administration. Yet Goldsmith, as just noted, writes as if this concern would be assuaged had the public only known about OLC “pushing” the law in order to approve harsh interrogation methods. Although a political realist, Goldsmith’s politics in these instances seem self-defeating.

Indeed, Goldsmith’s realism sometimes seems more a substitute for reality than an embrace of it. The overbroadness of the Torture Memo and its reliance on overly general (and questionable) legal rules disturbed Goldsmith. He sometimes writes, however, as if members of the public subscribed to a well-developed legal philosophy, like Goldsmith’s own, that placed the highest value on OLC’s reputation and that focused more on the theory of legal opinion writing than on interrogation methods like waterboarding. The result resembles a fun house mirror, with the public seemingly objecting to Yoo’s interrogation opinions because they revealed too much—to the public itself.

This admixture of realism and unreality suggests a variation on Plato’s allegory of the cave.299 Plato’s cave dwellers were chained, preventing them from turning and seeing behind them, where a fire that cast light and some objects were located. Only the objects’ shadows, projected by the fire onto the cave walls, were visible. The cave dwellers understandably confused shadows with reality, regarding the shadows as real. Free of chains and outside the cave, philosophers could see everything: the fire, the shadows, objects and the cave dwellers.

The shadows and the reality behind the interrogation memos have become now public knowledge, but Goldsmith ignores the reality. He merely wants to recast the shadows into a form more favorable to OLC’s reputation.

298 See, e.g., The Torture Memos, ST. PETERSBURG TIMES, June 13, 2004 available at

http://www. sptimes.com/2004/06/13/news_pf/Opinion/The_torture_memos.shtml (noting that the highest administration officials “sought ways to defend the use of abusive tactics on the part of American military personnel and intelligence agents”).

299 PLATO, THE REPUBLIC 220–51 (G.R.F. Ferrari ed., Tom Griffith trans., Cambridge Univ. Press 2000).


C. The Anti-Torture Chimera

Goldsmith, as noted in the Introduction, has been widely viewed as a defender of the rule of law and as an opponent of harsh interrogations.300 But Goldsmith never finds interrogation techniques illegal, and if “rule of law” means what anti-torture advocates think it does, Goldsmith was not a defender of the rule of law, either.

In part, the misunderstanding is fueled by Goldsmith’s actions as OLC director. The historic framework is binary, and the Bush administration is often viewed as pro-torture. Since Goldsmith fought the White House, withdrew the Torture Memo and resigned, one is inclined to place him on the opposite, anti-torture side.

Goldsmith’s failure to author an interrogation opinion at OLC has abetted this misimpression. Had Goldsmith formally upheld interrogations at OLC, it would be more difficult to assign him to the anti-torture camp. It would also be more difficult, as noted earlier, if The Terror Presidency had described the interrogation techniques approved by the Legal Standards and Qaeda Operative memos, both of which Goldsmith considered free of the Torture Memo’s flaws.

Events did not create the misunderstanding alone, however. Goldsmith’s writing fosters it. An overriding theme of The Terror Presidency is principled opposition to the Torture Memo, grounded in a general approach to law. Thus, on the surface, Goldsmith appears to echo the anti-torture critics who see blatant legal errors that they trace to fundamental misconceptions. But Goldsmith does not subscribe to the principles that animate opposition to torture. Nor do critics of torture have any reason to place a high value on OLC’s reputation, or to share Goldsmith’s realist and minimalist philosophy. Moreover, where anti-torture writers decry the Torture Memo as unprincipled, Goldsmith essentially faults it as too principled. Where torture critics strive to preserve the nation’s moral identity and commitment to the rule of law, Goldsmith simply does not recognize moral identity or an abstract rule of law as operational entities; in their place, we find OLC’s reputation and the President’s agenda.

Goldsmith does invoke “fidelity to law,”301 which sounds somewhat like “rule of law.” However, because fidelity to law seems to lack any internal component, there is no internal conceptual structure that torture conceivably could harm. Further, since Goldsmith’s approach to indeterminate legal issues is guided by the President’s agenda and by a

300 See supra note 43 and accompanying text. See also supra p. 162–63 (suggesting that

differences between Goldsmith and Yoo have been exaggerated). 301 GOLDSMITH, supra note 20, at 39.


balancing formula anchored to harsh interrogations, the idea of “fidelity to law” in such cases seems nearly meaningless.

When law is determinate and draws a definite line, fidelity to law presumably requires respect for that line. Goldsmith followed this precept in connection with the National Terrorism Surveillance Program, which he opposed.302 Yet, formal legalism of this kind is hardly incompatible with torture or other abuses, especially when the area of legal indeterminism is so large.303 Nor does Goldsmith adhere unwaveringly to the “fidelity to law” ideal (as evidenced by his belief that sometimes “legal sleight of hand” is required).304 Thus, “fidelity to law” is quite different from the “rule of law” of anti-torture arguments—it only sounds similar. Like “way of life,” fidelity to law is easy to confuse with a staple of anti-torture arguments.305 And, as in the earlier cases, an ersatz concept gets substituted for a real one.

Readers can misunderstand Goldsmith precisely because their internal mental map incorporates the historic pattern. “Way of life” and “fidelity to law” would signify very little without that. Thus, Goldsmith’s chimera becomes visible only against the backdrop of the historic debate about torture.

D. Manichaeism

Yet another consideration may have contributed to misunderstandings about Goldsmith. Abstract concepts have undergirded opposition to torture since the time of Beccaria and Voltaire. But this association, as noted in the Introduction, is not required by logic. Pro-torture arguments can rest on abstractions too—for example, on a theory about eradicating evil and the soul. Thus, one can imagine anti-torture arguments based on opposition to those abstractions and perhaps even on opposition to abstractions in general. And that would reverse the classic pattern of debate. Pro-torture arguments would become linked with abstractions, and anti-torture arguments on oppositions to abstractions.

Goldsmith perhaps believed that something like that was underway within the Bush administration. Discussing a magazine story about him, Goldsmith noted its unfortunate “Manichean tone” and the mistaken suggestion that his disagreements with other administration officials were

302 Id. at 39–41. 303 For an extreme example of formal legalism see RICHARD J. EVANS, THE THIRD REICH IN

POWER 1933–1939, 74–75 (2005) (describing objections by Nazi Ministry of Justice officials to transferring prisoners to concentration camps during their sentences—while having no objection to the Gestapo’s seizing prisoners and sending them to camps the moment that their sentences ended).

304 See supra pages 162–64. 305 For discussion, see supra text accompanying notes 276–78.


“struggles between the forces of good and evil.” 306 Yet, it is easy to read The Terror Presidency in just that way. For Goldsmith treats his differences with other officials as the result of fundamentally opposed approaches to law.

Goldsmith portrays Addington and Yoo as driven by the overriding principle of aggrandizing executive power in the name of the Constitution. They regarded “executive power” in Article II as an abstraction with substantial internal content, and from this, drew elaborate inferences about Presidential power.307 Goldsmith, on the other hand, is a “realist.” In that situation, Goldsmith’s opposition to abstractions might seem to mesh with opposition to torture. As already noted, Comey opposed the Combined Techniques Memo on the ground that no generalizations about combining interrogation methods should be made, rather, each that combination, and every detainee, were different and required individualized legal opinions.308

The possibility that Goldsmith viewed abuses as the consequence of abstractions may underlie The Terror Presidency’s anomalies, ironies and inconsistencies. It helps account for Goldsmith’s sense that his objections to the Torture Memo rested on deep principle—even though the principle was that OLC should generally avoid broad principles. It may also explain his apparent assumption that only good things come from realism, even though realism can easily justify torture and other abuses. Further, it makes understandable Goldsmith’s seeming conviction that OLC’s reputation was better served by his favoring the President’s agenda and applauding past legal “sleights of hand” than by the Torture Memo’s broad interpretations of the law.

Even if the historic pattern broke down inside the Bush administration —and it is not clear that it did—the classic anti-torture stance remained the best response. It was the public outcry over the Torture Memo, a reaction within the classic pattern, which finally prompted Goldsmith to withdraw the opinion.309 Moreover, a realistic philosophy should be assessed in light

306 GOLDSMITH, supra note 20, at 175. Goldsmith found it “unsettling and somewhat

embarrassing that so many people who detested the administration, and until the Newsweek article didn’t much like [him], were calling [him] a ‘hero.’” Id. at 174.

307 Vice President Cheney and David Addington subscribed to the idea of a “unitary executive,” a theory that yielded an extraordinarily powerful President. Id. at 85–86. In line with the theory, they were committed “to expanding presidential power,” Goldsmith writes, and “through their influence, President Bush and Alberto Gonzales” came to share the same commitment. Id. at 89. This belief was shared by John Yoo. According to Goldsmith, Yoo’s version of unitary executive theory held that “when the Constitution vested `the executive power’ in the President, it gave [the President] all of the military powers possessed by the King of England save those expressly given to Congress.” Id. at 97.

308 See supra text accompanying note 170. 309 GOLDSMITH, supra note 20, at 158.


of its real world effects, and in reality, Goldsmith’s realism proved compatible with the administration’s harshest interrogation methods. If those interrogation methods grew out of a revived, pro-torture conceptualism, it would be so much the worse for realism.

E. A New Leviathan

The cover of Leviathan, Thomas Hobbes’s 17th century classic of authoritarian philosophy, depicted the “sovereign” as a giant human form with a body comprised of tiny individual persons, like the paisley pattern in a fabric.310 The image captures Hobbes’s idea that the sovereign is, literally, the people.311 It also reduces the people to something like the sovereign’s “muscles and fibres.”

Although Goldsmith is hardly a political absolutist, there are suggestive parallels with Hobbes. Hobbes, like Goldsmith, opposed political abstractions. No principles were needed to control the sovereign’s relations with the people, Hobbes believed, because sovereign and the people were one.312 Hobbes completely eliminated any role or function for political principles by the act of concrete identification of sovereign and people.313

The Terror Presidency includes echoes of this idea. Recall Goldsmith’s claim that it is legitimate for OLC lawyers to advance the President’s agenda because the lawyers’ philosophical attunements are in line with the President’s.314 Goldsmith’s argument seems unconvincing, and perhaps the reason is that it relies on a Hobbesian identification of lawyers with the President, not on principles, to explain lawyers’ obligations to the administration. Elsewhere in the book, Goldsmith links OLC (and himself) to the intelligence community—the people of the executive branch—in an unusually intimate way, even if he does not quite identify the two.315 For that matter, linking national “way of life” with aggregated individual lives, as Goldsmith does in his balancing formula, is a close linguistic approximation to the image on the frontispiece of Leviathan.316 It, too, replaces abstract principles with concrete individuals.

310 THOMAS HOBBES, LEVIATHAN OR THE MATTER, FORME & POWER OF A COMMONWEALTH, ECCLESIASTICALL AND CIVILL (Cover) (A. R. Waller ed., Cambridge Univ. Press 1904) (1651).

311 Hobbes’s formulation is remarkably close to the modern idea that the people are sovereign, yet it means the exact opposite.

312 See generally HOBBES, supra note 310. 313 See generally id. 314 See discussion supra Part IV.B.2. 315 See, e.g., GOLDSMITH, supra note 20, at 162 (saying that one reason for his resignation is that

“many of the men and women who were asked to act on the edges of the law had lost faith in [him]”). 316 See supra note 310 and accompanying text.


Both Hobbes and Goldsmith give a personal cast to law. Hobbes regarded sovereign wishes as law; Goldsmith believes the President’s agenda legitimately enters into the meaning of law.317 Hobbes considered the sovereign above criticism; Goldsmith posits a fundamental imperative to minimize criticism of the executive branch.

There are also suggestive personal parallels with Hobbes. Although his authoritarian views supported royal absolutism, Hobbes dismayed the King’s supporters. The difficulties arose from Hobbes’s religious beliefs, which have been described as a “version of Christian theology designed to fit in with a materialist philosophy.”318 Because of that theology, ministers of the newly restored monarchy who once “admired Hobbes came to “detest[] him for his apparent treachery.”319

Goldsmith’s situation in the Bush administration was remarkably similar. He differed with Addington and Gonzales largely over questions of jurisprudence, as we have seen. But just as Hobbes’s materialism overshadowed his embrace of royal power, Goldsmith’s constitutional realism and minimalism seemingly outweigh his basic agreement with administration interrogation policies.

The result in both cases was ironic. Hobbes, a materialist, lost favor because of ideas about religion. And Goldsmith, a minimalist who eschewed broad principles, provoked deep disagreements over questions of legal philosophy.320

F. Misunderstanding and Confusion

Shortly after publication of The Terror Presidency, Goldsmith testified at Senate Judiciary Committee hearings. The subject was “Preserving the Rule of Law in the Fight Against Terrorism.”321 As framed by Senator Leahy, the committee Chair, the question was whether the nation could “maintain respect for the rule of law and our Constitution” during a “time of crises.”322

Committee members praised Goldsmith and his book. Senator Schumer, for example, “congratulate[d]” Goldsmith for his “attempts to uphold the rule of law” and “for doing something to return the

317 Compare GOLDSMITH, supra note 20, at 34–35 (explaining that “advice to the President needs

to have the political dimension clearly in view” and need not be “entirely neutral to the President’s agenda), with HOBBES, supra note 310.

318 RICHARD TUCK, HOBBES 29 (1989). 319 Id. at 33. 320 See generally supra Part IV.C. 321 Rule of Law Hearing, supra note 187. 322 Id. at 1 (statement of Sen. Leahy).


administration to the path of law.”323 Schumer added, “[o]f course, I know when you first went to . . . Harvard Law School, you were protested by the left, and now I guess they would admit that they were wrong, I hope.”324

Goldsmith’s account of the “rule of law” in his testimony centered on the executive branch, rather than the courts or the legislature—and it did so in part because of the president’s obligation “to keep the country safe.” Goldsmith began by noting that the rule of law was “something that I’ve thought a lot about in the last three or 4 years.”325 He continued:

The first institution that must be focused on,

obviously, in answering this question is the presidency . . . because the President, under the Constitution, has two duties that are relevant here. First, he has the duty to keep the country safe . . . . And the President also has a duty to take care that the laws are faithfully executed, so he has a duty to comply with the law.326

Goldsmith alluded to other branches, but the presidency always

remained on center stage. We must, he said, “acknowledge the difficult position that the executive branch is sometimes in. The . . . main lesson . . . over the last 6 years is that the institutions of our government have to work together to manage the problem. It’s nothing . . . that one institution alone can do.”327

On the subject of the “law,” which Goldsmith had been praised for defending, he observed:

[T]here’s the law and the need to comply with the law. In my experience, people . . . throughout the administration . . . did try to comply with the law. Some people had . . . different views of what the law required, [and] some people said that the Bush administration has been indifferent to law, but in my experience it’s been preoccupied with law. There are lawyers in every meeting related to counterterrorism policy.

The Department of Justice has probably issued more opinions related to this war than all of its opinions related to wars in the past. So at the same time that the

323 Id. at 21 (statements of Sen. Schumer). 324 Id. 325 Id. at 5 (statement of Jack Goldsmith). 326 Id. 327 Rule of Law Hearing, supra note 187, at 7.


administration is pushed to try to stop the attack, it finds itself bumping up against laws, lots of laws . . . .328

The substance of Goldsmith’s testimony, and much of his language,

echoed The Terror Presidency. Goldsmith spoke of “law” not as a living entity capable of a setting “path”—the phrase Senator Schumer had used329 —but as something material and almost physical. Law was a thing one might “bump[] up against,” that there could exist too much of, and that the President could “take care [of].”330

These exchanges with Goldsmith were marked by a profound misunderstanding. Senators regarded law as an abstraction with a complex internal life. Goldsmith saw it as a collection of rules and government lawyers. No one noticed the difference—not until Senator Kennedy posed a written question to Goldsmith.

Kennedy began by noting Goldsmith’s claim that “the administration has paid scrupulous attention to law.”331 Kennedy regarded this as “curious” because of other assertions in The Terror Presidency, such as: that “OLC was under heavy pressure to justify what the President wanted to do”; that “OLC ‘took shortcuts in its opinion writing procedures” that worked to “control outcomes in the opinions and minimize resistance to them”; that legal opinions were “deeply flawed”; that “opinion after opinion approve[ed] every aspect of the administration’s aggressive antiterrorism efforts”; and that the White House “used lawyers . . . as a sword to silence or discipline” those with qualms about the most problematic policies.332 For those reasons, Kennedy suggested that Goldsmith’s book actually supports the “exact opposite” conclusion, namely, “White House officials . . . consistently manipulated or ignored the law to suit their own ends.”333 “It is hard to come away from your book,” Kennedy wrote, “without concluding that [the Bush] . . . Administration is not committed to the rule of law.”334

Kennedy then asked whether Goldsmith believed “that the [Bush] Administration has been committed to the rule of law “not just as a means to justify their policies, but as a bedrock principle.”335 Goldsmith

328 Id. at 6. 329 Id. at 21. 330 Id. at 5–6. 331 Senator Edward M. Kennedy, Questions for the Record, in Rule of Law Hearing, supra note

187, at 40. 332 Id. at 41. 333 Id. 334 Id. 335 Id. at 42.


answered:

I do not agree with all of these characterizations of . . . The Terror Presidency. As I explained in the book, I believe the administration, far from being indifferent to law, was preoccupied with law. As I also explain in my book, I disagreed with some of the administration’s legal contentions and some of its approaches to ascertaining and complying with the law.336

In effect, Senator Kennedy asked how The Terror Presidency squared

with classic anti-torture conceptions of law. For his part, Goldsmith appeared to not understand the question. Kennedy’s questions were posed in terms of classic conceptions and Goldsmith’s answers were not. The “rule of law” that Kennedy invoked simply was not part of Goldsmith’s conceptual equipment. Goldsmith had created a chimera of law, but, as this exchange demonstrates, he sincerely believed in his creation. Like the cave dwellers, he was incapable of conceiving a reality beyond the shadows.

VI. CONCLUSION

The Department of Justice has exonerated the lawyers responsible for interrogation opinions. The Department’s conclusions, however, rest on implausible legal and ethical premises. Investigators considered it worthy of exoneration if an attorney avoided statements of legal principle in an opinion. Investigators also considered it ethically exonerating if an attorney wrote opinions calculated to minimize damage to the Department’s own reputation, should secret interrogations or secret legal opinions ever become public. Further, the Department did not believe that the ethics issue required a determination whether the interrogation opinions were legally accurate. Investigators exonerated an attorney who had reached obviously “counterintuitive” conclusions, seeing no need to determine whether those counterintuitive conclusions were mistaken.

In a preliminary report, investigators found Jay Bybee and John Yoo—who had incorporated broad legal principles into their opinions, and paid no attention to the possible public reaction—guilty of professional misconduct. That conclusion was overturned by a senior Justice Department reviewer, but the reviewer questioned none of the investigation’s basic premises. In fact, the reviewer also faulted Bybee and Yoo for broad statements of legal principle and lack of public relations

336 Id.


savvy. They were exonerated, however, largely on the ground of legal indeterminacy. The review suggested that, no matter how suspect Bybee’s and Yoo’s conclusions were about interrogation, the conclusion by investigators that Bybee and Yoo had performed unprofessionally was just as suspect.

These findings were remarkable. The ethical issue appeared to be whether lawyers upheld illegal torture by repeatedly misinterpreting the law. Yet the investigators: i) avoided the question of whether the lawyers were wrong; ii) posited an ethical imperative to produce legal opinions calculated to minimize damage to the Department’s reputation, regardless of the techniques being upheld; iii) disdained broad legal principles, both as an ethical matter and as a matter of legal craft; and iv) suggested enormous skepticism about anyone’s capacity to discover the meaning of legal rules. The result bordered on surreal. It was as if the Department inhabited a universe of ethics and law subtly, yet also bizarrely, different from our own.

The standard framework for thinking about torture and executive power—and about their relationship to law—relies on abstractions like personal and national identity and the rule of the law. However, an alternate framework has recently emerged, and it accounts for the Department’s anomalous conclusions. The alternative, which is most closely associated with Jack Goldsmith, both resembles the standard framework and manages to stand the standard framework on its head.

The alternative is radically anti-abstraction, highly realistic in both politics and jurisprudence, and strongly slanted toward Presidential power. It reduces abstractions like humanity and the rule of law to nothing more than potential political flash points—public nerves endings that could trigger a reaction by the body politic against the executive branch. These attributes, which the Article describes in detail, sharply contrast with those of the standard framework.

The alternative is a chimera, I argue, but a powerful one that has become becoming increasingly influential. It led the Department of Justice to miss the point of its own ethics investigation. It prompted Senators to hail Goldsmith as the champion of a conception of the rule of law that, at a Congressional hearing, he did not even recognize.

We should acknowledge this legal chimera for what it is. Today, it underlies the emerging idea of an executive branch limited only by political considerations and a body of executive branch constitutional law crafted by the executive itself—a body of law developed (in Goldsmith’s words) by “neutral” and “independent” executive branch lawyers who regard the President’s philosophical “attunements” as a significant component in legal interpretation and who are willing to “push” the law to further the President’s political “agenda.” If this chimera continues to be mistaken for our historic conceptions, it will confound our future, too.

Lawyers, Interrogations and the Historic Framework of ...€¦ · 2010] LAWYERS, INTERROGATIONS AND THE HISTORIC FRAMEWORK 103 personalities, or they are sources of information about

Documents